TI to Overhaul Fab for 65 Nanometer and 45 Nanometer Production

Texas Instruments has announced that it plans to overhaul its fab for 65 nanometer and 45 nanometer production for analog and embedded processors once the purchase closes. This will be TI's fourth fab of this type, with the company currently having DMOS 6, RFAB 1, and soon-to-be-completed RFAB 2. The new fab is expected to enable faster and more efficient manufacturing processes for TI's customers.

The process of developing and manufacturing semiconductors is a complex and ongoing process that requires significant investment in research and development. TI has been investing heavily in its fabs to stay ahead of the competition and meet the growing demands of its customers. The company's commitment to innovation and quality will continue to be reflected in its products and services.

The Wafer Agreement Between Intel and Micron Expired

Intel and Micron have recently seen their wafer agreement expire, marking an important milestone for both companies. This expiration has significant implications for the future of 3D crosspoint memory, a technology that Intel plans to continue developing and manufacturing. However, with Micron exiting the business, Intel will no longer be able to purchase 3D crosspoint memory from its former partner.

Intel's decision to abandon its partnership with Micron is not surprising, given the company's long-term commitment to innovation and quality. While the loss of this partnership may present some challenges for Intel, the company remains confident in its ability to develop and manufacture new technologies that meet the needs of its customers.

The Resurrection of the 3dfx Voodoo 5 6000

A modder, Anthony, has brought back another GPU from the depths of history - the 3dfx Voodoo 5 6000. This is not the first time Anthony's work has been featured on Cameras Nexus, as his previous creation, a combination of a 16 megabyte Voodoo 3 3500 and a 4 megabyte PowerVR device, was also showcased in a previous video.

Anthony's latest creation is a custom-built GPU that combines both retro GPUs atop a black PCB. The Voodoo 3 3500 runs at an impressive 183 megahertz, while the PCX2 runs at a more modest 66 megahertz with four megabytes of SDRAM. Both GPUs make use of passive heat sinks and feature a special jumper for overclocking.

Anthony's work is a testament to the creativity and ingenuity of modders and engineers who continue to push the boundaries of what is possible with vintage technology. While not all of these projects may be practical or functional, they serve as a reminder that even the most obscure technologies can still inspire new ideas and innovations.

Samsung and Synopsys Collaborate on 3 Nanometer GAA Process

Samsung and Synopsys have recently announced their collaboration on a groundbreaking 3 nanometer GAA process. This technology is based on what Samsung calls "three nanometers" and uses Gate-All-Around Transistors (GAA Phets or GATE). These types of transistors are largely the future for scaling silicon and have significant implications for the industry.

The GAA transistor structure marks a key inflection point in process technology advancement that is critical in maintaining the scaling trajectories needed for the next wave of hyperscale innovation. Synopsys has played a crucial role in developing this technology, providing its expertise in design verification IP and software to support Samsung's efforts.

Samsung's customers will have access to Synopsys' technology files through Samsung Foundry, meaning that clients of Samsung's foundry services will also gain access to similar designs. This collaboration between Samsung and Synopsys is a testament to the strength of their partnership and commitment to innovation in the field of semiconductor manufacturing.

"WEBVTTKind: captionsLanguage: enhey everyone welcome back to another hardware news recap for the week in this one we'll be talking about intel getting on the list for tsmt's three nanometer process alongside apple we'll also be going over amd updating dlss and andy has officially shipped an update for nvidia's dlss the update's called fsr we'll be going over that uh more intel gpu details microsoft panicking in response to feedback on windows 11 and back back pedaling so hard it might trip over we'll also be going over used mining gpus beginning to service on the market and a couple of industry news stories for the past week before that this video is brought to you by be quiet and the silent base 802 case the silent base 802 got high accolades in our review for its high build quality and its versatility in both sound focused and airflow focused built the 802 comes with swappable mesh panels or noise damped panels so you have options for either approach the silent base 802 case is able to fit larger builds as well without being overbearing and it stands out for its mechanics quality and assembly quality learn more about be quiet's new case at the link in the description below so first up just a quick section on what we're working on right now and you can check the time bars over there if you'd like to skip it but some stuff we're working on first of all there was a midweek news recap that we ran a couple days ago and that talked about evga's x570 motherboards coming out we have a lot of new information in there that wasn't previously known because we spoke with kingpin from evga and with other reps at evga to get some details on the boards the company's working on so if you missed that you might be interested in it just kind of interesting because ebj is moving some of its intel only stuff to amd uh we also covered a ryzen software utility for overclocking in that that is called project hydra it's from one usmas the person who worked on things like the ryzen dram calculator for example so we've got information from him in that as well if you're curious now as for stuff we're working on beyond that uh we have some more intel dg1 content coming up around the channel already we've got two more videos on that that are coming out we are working on a 7000 d review that case is out now it's a corsair case it follows up the 4000 and the 5000d now the 7000d not to spoil it but but 7000d it has 2 000 more d's than the 5000 d actually 3 000 more d's than the 4000 d obviously that means something so uh it's probably better that's that's what we think that probably means but we're going to figure it out for you and we'll post it in a case for you coming up probably in the next week or so some videos out there already about it but we're working through the normal reviews process as we always do we also have some more low end cooler benchmarks coming up versus the 68 watt heat load we just started using that for the noctua nhp1 passive cooler and that finally pushed us to get into gear on a 3600 5600x style heat load to test coolers like amd stock coolers or other lower end tower coolers so that's coming up and finally there's a ram timings behind the scenes video up on patreon on patreon.comgamersnexus where andrew and patrick stone talk through some of the work they've been doing on our refreshed ram timing series with the uh multiple parts and that's going to be a while in the making yet still because we're doing a lot of animations for it so that's on patreon and then if you want to grab one of these the gn wireframe mouse mats they are on the store on store.camerasexus.net they're in stock and shipping now we ordered a huge amount of them we did that last the last two or three times as well and they always sell out really fast i think maybe we might last time we were doing pretty well we got about a month or two of inventory in before we sold out of them i think we might have done even better this time but they still sell through quickly uh thank you for all the support for buying them we're glad that you get something that is a a high quality usable product that also supports us so everybody wins thanks for picking them up and to those of you who've tweeted us the photos of your setups everybody in our audience apparently has a better setup than i do so uh they all look great but thank you for the support and you grab one on store at access.net let's get started with the news items so for this week the first one we'll talk about is the three nanometer process uh tsmc wishlist from intel and apple up first this is according to a report from nikai asia review where intel and apple are listed as the earliest adopters of the three nanometer process so the report claims that intel and apple are already testing and validating their designs with tsmc and it cites sources quote briefed on the matter we already discussed rumors previously that apple snatched up a sizable portion of tsmc's initial capacity to no one's surprise really but this is the first we're hearing about intel claiming any as for what products we'll be using the three nanometer process from tsmc at least early on it looks like apple's ipad line will probably be the first from apple to use three nanometer silicon from tsmc and as far as intel goes they have two product lines that are rumored in the works for it one is on notebooks and one is on data center those are both fairly high volume businesses for intel or at least on the data center side high revenue if not high volume and so they are keystones for the company's income every every quarter now more interesting maybe is that tsmc is already allocating more capacity for intel than apple at least right now which would be a surprise we're a little bit reluctant to believe that apple is a massive customer for tsmc so we'll see we'll see how that pans out intel's a big company as well but i mean apple so uh beyond that what we know currently comes from a we've got a quote from nikai asia review actually it says quote currently the chip volume plan for intel is more than that of apple's ipad using the three nanometer process the media outlet reached out to tsmc apple and intel for comments intel only reiterated its partnership with tsmc for its 2023 products while tsmc just declined to comment as taiwan semiconductor doesn't comment on its customers products and that's really how most these companies act apple for its part did not respond to the request for a comment up next amd updated nvidia dlss amd shipped an update called fsr a little while ago we uh we benchmarked it so fidelityfx super resolution as we said in the fidelity fx super resolution sort of review and quality image quality analysis it's not meant to be a dlss direct competitor but they sort of market it in the same category of software designed for gpus to improve performance and it works differently and we talk about that in review if you're curious about how it works differently we went over that in a whole section but the point is that even though it's not meant to be a sort of direct competitor to dlss that's how it's being received in a lot of ways and so as you would expect nvidia has now updated dlss in response it would seem at least fsr and video would probably claim that it's been working on this update uh but there's a few things that make it seem a little bit more like a response than like something that was already in the works so andy is obviously not literally updating dlss but the change from nvidia is that it introduces a new setting called ultra quality which is something that it was specifically lacking and that amd had in it now if this seems arbitrary and like just the name of a setting that is non-standardized and doesn't actually need to relate to anything or be used by any company then then you're correct and thinking that it seems that way but that's not going to stop companies from marketing against each other by using a name with ultra in it which of course establishes a a hierarchy of being the best and then using it for their own product so amd's fsr followed nvidia's naming scheme and resolution scaling actually with a performance mode a balanced mode and quality mode that equaled nvidia's dlss scale factors and we could put a chart of one of those on the screen nvidia differentiated itself with its own ultra performance mode and all this of course came out before fsr it was the first with a feature like this at least to this degree and ultra performance it's much lower quality and higher performance than amd's fsr but because nvidia can train its solution it has some advantages that make it able to reduce the pixel data input and perhaps get an image that's still usable out the other end either way amd is differentiating itself with ultra quality or at least it was this is a scale factor closer to 1.0 than the others at around 77 input resolution and not to be outdone by something called ultra quality nvidia decided to now work on its own ultra quality mode we'll see if that's the final name for it frankly we're surprised that nvidia didn't decide to name it something like thai quality or uh maybe super quality ultra extreme quality it's a lot of things that would seem better than what amd's doing but no no it's it's ultra quality right now so maybe the industry is at a point of agreeing a little bit on names for ones that would be a nice thing with usbif and intel for example the discovery of this new mode was made by reinhardtovic over on nvidia's subreddit where the user spotted nvidia and a change noted in unreal engine 5's dlss documentation at this time specific details about the scaling and resolution support are unknown or unclear dlss version 2.2.9.0 though should coincide with this rollout and nvidia is actively updating dlss as it has been up next broadcom mediatek and marvel apparently supporting nvidia is a buyout of arm so there's there's a list of armed licensees who we're going to use the word seemingly here support the nvidia arm acquisition it's also a growing list of companies that oppose the nvidia arm acquisition mostly for the reasons you would expect like not being an independent company anymore that is potentially part of a competitor so there's no shortage of licensees to oppose the deal qualcomm most recently is the one that was in the news for opposing the nvidia arm acquisition and qualcomm suggested as we said in the news video last week that arm basically go public and or have a consortium of of technology company supporters instead who all have a stake in the company this time broadcom mediatek and mar-vell all came out to approve of the deal this is according to the times paper in the uk that says these companies have broken rank to support nvidia as the regulatory process heats up in the uk the report from the uk times is paywalled but it doesn't seem there's much in the way of statements or explanations anyway for the newfound support for the 40 billion purchase of arm meanwhile the uk's secretary of state for digital culture media and sport recently stepped in to block the deal on the grounds of national security concerns and the uk's competition and markets authority or cma is currently preparing a report on the potential anti-trust and national security implications of such a deal that report is due out by the end of july so we'll know more then and it should serve as a way to inform how the uk's regulatory bodies approach the deal additionally the nvidia arm deal has hit expected hurdles worldwide including the us and china as always we'll continue to watch these updates closely because this story has major implications for the entire industry intel had a presence at the 2021 international supercomputing conference and this time it was talking about its ponte vecchio gpus and it's sapphire rapids cpus so this includes some information on its next generation xeon scalable cpus and the xe hpc based ponte vecchio intel sapphire rapids is an increasingly important platform and one that intel powered on just over a year ago sapphire rapids will be the platform for its biggest supercomputer victory to date which is the aurora supercomputer and it'll also be the vehicle for intel to debut its own ddr5 support as well as pcie gen 5 and cxl 1.1 support of course among other technologies now intel has announced a new fold for sapphire rapids there will be variants that come with onboard hbm or high bandwidth memory intel sapphire rapids with hbm will of course be of particular interest to those in its hpc markets and intel and its press release on its website was quick to quote several professionals from national laboratories such as argonne national laboratory where aurora will live and los alamos national labs which is the home of the crossroads supercomputer until at this point hasn't told us what type of hbm it's going to be using there are several out there's hbm hbm2 hbm2e etc uh and it also noted that the cpus could be used with or without system memory so that's actually pretty interesting and that's the aspect we're the most interested in seeing if that ever sort of comes down the chain to consumer stuff at some point in the distant future in addition to hbm intel also officially confirmed its new ai acceleration engine that will appear on sapphire rapids and that is called advanced matrix extensions or amx moving on to ponte vecchio intel announced a few new details intel's ponte vecchio is based on intel's xe hpc silicon and was successfully powered on earlier this year intel now states that ponte vecchio is currently in the process of system validation and it'll arrive in an ocp accelerator module or oam form factor intel will also offer ponte vecchio in subsystems offering either four or eight oam packages of course the first silicon for it is destined for the aurora supercomputer but this is the first time intel has detailed its plans for ponte vecchio outside of supercomputers as expected there's been a deluge of information as microsoft took the lid off of its windows 11 announcement and the ensuing uh what's known in the technical world as big oof that followed and this was because of microsoft's cpu and tpm requirements which stirred up no shortage of both confusion and criticism initially microsoft stated that cpu support would begin with intel's eighth generation or eighth series that would be 8700k for example 8600k and also that would begin with amd's ryzen 2000 series ryzen 1000 the initial came out around 2017 or so and uh it's also around when the eighth series came out from intel now this in itself didn't disqualify the older cpus if you look at windows 10 for example there are plenty of cpus that are technically unsupported that support windows 10 or run windows 10 just fine but it was no less disappointing to see microsoft so hyper focused on newer hardware and of course you you think outside the consumer space where people are a little more conscious about what they're spending using you look at the corporate spaces and most of the big companies are going to play exactly by the book and follow the specs that it officially supports so that does start to limit what's usable with windows 11 in those environments now in an effort to smooth over the hardware concerns among others microsoft stated in a blog post that it's working to validate systems using intel's 7th gen cpus which came out about january of the same year as the 8th gen cpus and which are functionally the same as intel's 6th and 4th gen cpus and we don't talk about 5. so they're doing a little bit better they're also adding ryzen 1000 cpu official support but that's where it stops for both of them additionally microsoft's received no shortage of criticism for its laughable pc health check app which offered literally zero insight as to why a machine was or wasn't supported in fact the open source app why not win 11 proved more useful online as such microsoft announced that it has pulled the app from downloads citing the need to address user feedback namely that it wants to make the tool do something it said quote based on the feedback so far we acknowledge that it was not fully prepared to share the level of detail or accuracy you expected from us on why a windows 10 pc doesn't meet upgrade requirements there was also some level of confusion around windows 11 would roll out to be clear windows 11 has a free upgrade path to current and eligible windows 10 users though those roll outs won't happen until some point during the first half of 2022 microsoft will begin rolling out certified win 11 devices in late 2021 they say around october and it noted that the rollout would take place across several months microsoft just released the first preview build of windows 11 to windows insiders where it expects to collect feedback in a number of areas because obviously microsoft noted that during this time the feedback will inform its decision on the minimum hardware requirements going forward up next team group is the first memory vendor that we've seen anyway out the door with available ddr5 modules despite the dearth of cpus and motherboards actually supporting the the new memory the emergent memory team groups elite ddr5 4800 kit has surfaced first and it has listings on both amazon and newegg right now both of the retailers do show this kit out of stock or unavailable so whether you can get it's a different story but team group's elite ddr5 4800 kit is 32 gigabytes 16x2 with two 16 gigabyte udem modules rated at 4800 mega transfers per second as expected the dimms come with ondi ecc and an onboard pmic team group's first ddr5 kit also has timings of cl 40 40 40 77. it's got a cast latency of 40 and an operating voltage of 1.1 volts both retail listings show the kit price at 311 u.s and it looks like they're expected to ship at some point in july the first consumer platform to support ddr5 will be intel's alder lake that platform is currently not a well it's officially announced until did show it in its disaster of a keynote again uh but it's rumored for an ounce late 2021 we don't have much information about it beyond that around the other on the release date and it'll be a while before you can actually really use ddr5 in the consumer market anyway so ddr5 will also take a while to scale beyond the best of ddr4 as a reminder the switchover period from when stuff changes like say from ddr3 to ddr4 that period of time the new memory is not advanced to the point where if you have really high-end existing memories ddr4 uh versus brand new what will eventually be low end ddr5 there's not normally that bigger performance gap there and sometimes it favors the older kit instead it just depends on the architecture of the cpu and the platform you're buying for of course thanks steve up next the usgpu market changing so as a result of china's crackdown on cryptocurrency the first wave of used mining gpus at least on mass has appeared online for the aftermarket and this was reported by pc gamer where screenshots posted to the taiwanese bulletin board ptt was we're showing a growing count of nvidia and amd gpus for sale and the gpus range from nvidia's rtx 30 series and amd's rx 6000 series at the newer side of things two older cars like of course and these polaris based cars that'd be the rx 400 500 series ethereum miners have been especially fond of the polaris based cards in the past so it's not really much of a surprise that they're starting to pop up on ptt according to hkepc the used price for an rtx 3070 is about 3120 hong kong dollars which works out to around 400 usd other cards are priced similarly the rtx 3060 ti for example is around 309 usd the 3060 non-ti is 244 however the cards are also being sold in bulk and they are also used mining carts which means they've likely been in service 24 7. it's the bulk aspect though that'll make it harder to deal with that said this is the latest of several developments that could mean gpu prices and availability are coming back down to earth we recently reported that asrock was optimistic about gpu price and stock uh especially as it expects amd's supply of gpus will improve during the second quarter or at this point rolling to towards third quarter of 2021. it's not just gpu shortages as in the the physical piece of silicon that were a problem though it's things like substrates and components too azeroth commented on that last week we talked about it as well where it said it expects substrate shortages to improve later this year uh now moreover popular coins like ethereum and bitcoin at least recently as of this film changes all the time have dropped in profitability this also affects the uh the dumping of used mining cards on the market and how fast that's happening ethereum's moving eventually in theory to a proof of stake model which moving away from proof of work will also be beneficial to gpu supply so things might be lining up in a way that's beneficial for those of you who still need to get in on a gpu even if it is used and it's just a matter of of well how the cryptocurrencies behave at this point because they seem completely erratic and difficult to predict so who knows we'll see what happens uh micron will sell its 3d cross point fab to texas instruments texas instruments of course also known as thai micron just announced that it would sell off its lehigh utah fab previously when it said it would exit the 3d crosspoint memory business and micron intended to have the fab sold by end of 2021 although at that time it disclosed no details on potential sale options micron recently announced that its 3d crosspoint memory fab will be sold to texas instruments micron states that the economic value of the sale is 1.5 billion dollars with 900 million coming from ti for the purchase of the facilities and the remaining 600 million coming from secondary sales like fab tooling and assets micron's press release didn't state much of what texas instruments plans to do with the fab not really much of a surprise there other than to deploy its own technologies ti will be can't say that without thinking of tie now ti will also attempt to retain all the fab's existing employees it says once the transaction closes it also said in its own press release that it plans to overhaul the fab for 65 nanometer and 45 nanometer production for analog and embedded processors once the purchase closes this will be ti's fourth fab of this type ti currently has dmos 6 rfab 1 and the soon to be completed rfab 2 way for fabrication facilities worthy of note is just how long intel will be able to buy 3d crosspoint memory from micron 4 at this point the wafer agreement between the two parties expires at the end of 2021 and with this sale and microns exit from the business a new agreement will obviously not be possible intel will have to set up its own lines for any products using octane memory in the future if it decides to keep making it although a lot of that starts getting into data center and enterprise markets that we don't cover too closely up next the same model responsible for the resurrection of the 3dfx voodoo 5 6000 that reported on previously has brought back another gpu from the debt these are fun projects they're really cool it's not like you can do a whole lot with it other than think that it's cool but that's nice to have every now and then too so the modder we speak of is anthony i'm not gonna try sorry we'll just we'll put the name on the screen my apologies and i was documented by a youtube video from michael dale which is a name that i can say i think the latest creation is a combination of a 16 megabyte voodoo 3 3500 and a 4 megabyte power vr device that's the pc x2 the design is a custom creation from we're going to go with that normally i use the last name but we're going to go with anthony who has combined both of the aforementioned retro gpus atop a black pcb which looks great by the way and the voodoo 3500 runs at 183 megahertz that's right get in line for this one it's gonna this will be the the gpu of the future voodoo i i think they're going to be the one nvidia is probably not long for this world so the 183 megahertz gpu has access to 16 megabytes of sd ram via eight sdram packages the pcx2 runs at 66 megahertz with four megabytes of sdram the board also features a pass-through port and circuitry to allow for another graphics card to be looped in and the board also features a special jumper for overclocking the pcx2 to 80 megahertz both gpus make use of passive heat sinks although it seems active cooling is recommended depending on the clock speeds the board also uses a standard pci slot rather than a pcie slot and in all seriousness if someone could post a guide on how to say that name i'd really appreciate the guide because it'll probably come up again tried using google i don't trust the results i got but let me know how to say it we'll do that next time uh next time we come across one of these awesome mods by anthony all right samsung and synopsis collaborating on three nanometer gaa process this will be our last story for this week samsung and synopsys announced that through their collaboration samsung has been able to achieve first pass silicon for an soc using samsung's 3gaa process samsung's three gaa process is based on what samsung is calling three nanometers and uses gaa phets or gate all-around transistors these types of transistors are largely the future for scaling silicon and we've discussed them numerous times and news stories recently neither company went into details on the gaa transistors being used samsung at this point has long been working on a proprietary mbc phet which is a multi-bridge channel fed and it's based on nanosheets but like other gaa fats nanosheets will allow for tuning performance or power consumption by augmenting the nanosheet width there's a quote here as well the gaa transistor structure marks a key inflection point in process technology advancement that's critical in maintaining the scaling trajectories needed for the next wave of hyperscale innovation by synopsis in addition to this tape out samsung's customers will have access to synopsis technology files through samsung foundry so presumably samsung foundry clients will also have access to similar designs and synopsis if you don't know is a silicon design company that's focused on design verification ip and software that's it for the news this week thanks as always for watching you can subscribe for more go check out the channel for some of our follow-up dg1 standalone video card coverage uh or you can go to store.camerasnexus.net to grab one of our mouse mats or patreon.com to see that new behind-the-scenes video thanks for watching we'll see you all next timehey everyone welcome back to another hardware news recap for the week in this one we'll be talking about intel getting on the list for tsmt's three nanometer process alongside apple we'll also be going over amd updating dlss and andy has officially shipped an update for nvidia's dlss the update's called fsr we'll be going over that uh more intel gpu details microsoft panicking in response to feedback on windows 11 and back back pedaling so hard it might trip over we'll also be going over used mining gpus beginning to service on the market and a couple of industry news stories for the past week before that this video is brought to you by be quiet and the silent base 802 case the silent base 802 got high accolades in our review for its high build quality and its versatility in both sound focused and airflow focused built the 802 comes with swappable mesh panels or noise damped panels so you have options for either approach the silent base 802 case is able to fit larger builds as well without being overbearing and it stands out for its mechanics quality and assembly quality learn more about be quiet's new case at the link in the description below so first up just a quick section on what we're working on right now and you can check the time bars over there if you'd like to skip it but some stuff we're working on first of all there was a midweek news recap that we ran a couple days ago and that talked about evga's x570 motherboards coming out we have a lot of new information in there that wasn't previously known because we spoke with kingpin from evga and with other reps at evga to get some details on the boards the company's working on so if you missed that you might be interested in it just kind of interesting because ebj is moving some of its intel only stuff to amd uh we also covered a ryzen software utility for overclocking in that that is called project hydra it's from one usmas the person who worked on things like the ryzen dram calculator for example so we've got information from him in that as well if you're curious now as for stuff we're working on beyond that uh we have some more intel dg1 content coming up around the channel already we've got two more videos on that that are coming out we are working on a 7000 d review that case is out now it's a corsair case it follows up the 4000 and the 5000d now the 7000d not to spoil it but but 7000d it has 2 000 more d's than the 5000 d actually 3 000 more d's than the 4000 d obviously that means something so uh it's probably better that's that's what we think that probably means but we're going to figure it out for you and we'll post it in a case for you coming up probably in the next week or so some videos out there already about it but we're working through the normal reviews process as we always do we also have some more low end cooler benchmarks coming up versus the 68 watt heat load we just started using that for the noctua nhp1 passive cooler and that finally pushed us to get into gear on a 3600 5600x style heat load to test coolers like amd stock coolers or other lower end tower coolers so that's coming up and finally there's a ram timings behind the scenes video up on patreon on patreon.comgamersnexus where andrew and patrick stone talk through some of the work they've been doing on our refreshed ram timing series with the uh multiple parts and that's going to be a while in the making yet still because we're doing a lot of animations for it so that's on patreon and then if you want to grab one of these the gn wireframe mouse mats they are on the store on store.camerasexus.net they're in stock and shipping now we ordered a huge amount of them we did that last the last two or three times as well and they always sell out really fast i think maybe we might last time we were doing pretty well we got about a month or two of inventory in before we sold out of them i think we might have done even better this time but they still sell through quickly uh thank you for all the support for buying them we're glad that you get something that is a a high quality usable product that also supports us so everybody wins thanks for picking them up and to those of you who've tweeted us the photos of your setups everybody in our audience apparently has a better setup than i do so uh they all look great but thank you for the support and you grab one on store at access.net let's get started with the news items so for this week the first one we'll talk about is the three nanometer process uh tsmc wishlist from intel and apple up first this is according to a report from nikai asia review where intel and apple are listed as the earliest adopters of the three nanometer process so the report claims that intel and apple are already testing and validating their designs with tsmc and it cites sources quote briefed on the matter we already discussed rumors previously that apple snatched up a sizable portion of tsmc's initial capacity to no one's surprise really but this is the first we're hearing about intel claiming any as for what products we'll be using the three nanometer process from tsmc at least early on it looks like apple's ipad line will probably be the first from apple to use three nanometer silicon from tsmc and as far as intel goes they have two product lines that are rumored in the works for it one is on notebooks and one is on data center those are both fairly high volume businesses for intel or at least on the data center side high revenue if not high volume and so they are keystones for the company's income every every quarter now more interesting maybe is that tsmc is already allocating more capacity for intel than apple at least right now which would be a surprise we're a little bit reluctant to believe that apple is a massive customer for tsmc so we'll see we'll see how that pans out intel's a big company as well but i mean apple so uh beyond that what we know currently comes from a we've got a quote from nikai asia review actually it says quote currently the chip volume plan for intel is more than that of apple's ipad using the three nanometer process the media outlet reached out to tsmc apple and intel for comments intel only reiterated its partnership with tsmc for its 2023 products while tsmc just declined to comment as taiwan semiconductor doesn't comment on its customers products and that's really how most these companies act apple for its part did not respond to the request for a comment up next amd updated nvidia dlss amd shipped an update called fsr a little while ago we uh we benchmarked it so fidelityfx super resolution as we said in the fidelity fx super resolution sort of review and quality image quality analysis it's not meant to be a dlss direct competitor but they sort of market it in the same category of software designed for gpus to improve performance and it works differently and we talk about that in review if you're curious about how it works differently we went over that in a whole section but the point is that even though it's not meant to be a sort of direct competitor to dlss that's how it's being received in a lot of ways and so as you would expect nvidia has now updated dlss in response it would seem at least fsr and video would probably claim that it's been working on this update uh but there's a few things that make it seem a little bit more like a response than like something that was already in the works so andy is obviously not literally updating dlss but the change from nvidia is that it introduces a new setting called ultra quality which is something that it was specifically lacking and that amd had in it now if this seems arbitrary and like just the name of a setting that is non-standardized and doesn't actually need to relate to anything or be used by any company then then you're correct and thinking that it seems that way but that's not going to stop companies from marketing against each other by using a name with ultra in it which of course establishes a a hierarchy of being the best and then using it for their own product so amd's fsr followed nvidia's naming scheme and resolution scaling actually with a performance mode a balanced mode and quality mode that equaled nvidia's dlss scale factors and we could put a chart of one of those on the screen nvidia differentiated itself with its own ultra performance mode and all this of course came out before fsr it was the first with a feature like this at least to this degree and ultra performance it's much lower quality and higher performance than amd's fsr but because nvidia can train its solution it has some advantages that make it able to reduce the pixel data input and perhaps get an image that's still usable out the other end either way amd is differentiating itself with ultra quality or at least it was this is a scale factor closer to 1.0 than the others at around 77 input resolution and not to be outdone by something called ultra quality nvidia decided to now work on its own ultra quality mode we'll see if that's the final name for it frankly we're surprised that nvidia didn't decide to name it something like thai quality or uh maybe super quality ultra extreme quality it's a lot of things that would seem better than what amd's doing but no no it's it's ultra quality right now so maybe the industry is at a point of agreeing a little bit on names for ones that would be a nice thing with usbif and intel for example the discovery of this new mode was made by reinhardtovic over on nvidia's subreddit where the user spotted nvidia and a change noted in unreal engine 5's dlss documentation at this time specific details about the scaling and resolution support are unknown or unclear dlss version 2.2.9.0 though should coincide with this rollout and nvidia is actively updating dlss as it has been up next broadcom mediatek and marvel apparently supporting nvidia is a buyout of arm so there's there's a list of armed licensees who we're going to use the word seemingly here support the nvidia arm acquisition it's also a growing list of companies that oppose the nvidia arm acquisition mostly for the reasons you would expect like not being an independent company anymore that is potentially part of a competitor so there's no shortage of licensees to oppose the deal qualcomm most recently is the one that was in the news for opposing the nvidia arm acquisition and qualcomm suggested as we said in the news video last week that arm basically go public and or have a consortium of of technology company supporters instead who all have a stake in the company this time broadcom mediatek and mar-vell all came out to approve of the deal this is according to the times paper in the uk that says these companies have broken rank to support nvidia as the regulatory process heats up in the uk the report from the uk times is paywalled but it doesn't seem there's much in the way of statements or explanations anyway for the newfound support for the 40 billion purchase of arm meanwhile the uk's secretary of state for digital culture media and sport recently stepped in to block the deal on the grounds of national security concerns and the uk's competition and markets authority or cma is currently preparing a report on the potential anti-trust and national security implications of such a deal that report is due out by the end of july so we'll know more then and it should serve as a way to inform how the uk's regulatory bodies approach the deal additionally the nvidia arm deal has hit expected hurdles worldwide including the us and china as always we'll continue to watch these updates closely because this story has major implications for the entire industry intel had a presence at the 2021 international supercomputing conference and this time it was talking about its ponte vecchio gpus and it's sapphire rapids cpus so this includes some information on its next generation xeon scalable cpus and the xe hpc based ponte vecchio intel sapphire rapids is an increasingly important platform and one that intel powered on just over a year ago sapphire rapids will be the platform for its biggest supercomputer victory to date which is the aurora supercomputer and it'll also be the vehicle for intel to debut its own ddr5 support as well as pcie gen 5 and cxl 1.1 support of course among other technologies now intel has announced a new fold for sapphire rapids there will be variants that come with onboard hbm or high bandwidth memory intel sapphire rapids with hbm will of course be of particular interest to those in its hpc markets and intel and its press release on its website was quick to quote several professionals from national laboratories such as argonne national laboratory where aurora will live and los alamos national labs which is the home of the crossroads supercomputer until at this point hasn't told us what type of hbm it's going to be using there are several out there's hbm hbm2 hbm2e etc uh and it also noted that the cpus could be used with or without system memory so that's actually pretty interesting and that's the aspect we're the most interested in seeing if that ever sort of comes down the chain to consumer stuff at some point in the distant future in addition to hbm intel also officially confirmed its new ai acceleration engine that will appear on sapphire rapids and that is called advanced matrix extensions or amx moving on to ponte vecchio intel announced a few new details intel's ponte vecchio is based on intel's xe hpc silicon and was successfully powered on earlier this year intel now states that ponte vecchio is currently in the process of system validation and it'll arrive in an ocp accelerator module or oam form factor intel will also offer ponte vecchio in subsystems offering either four or eight oam packages of course the first silicon for it is destined for the aurora supercomputer but this is the first time intel has detailed its plans for ponte vecchio outside of supercomputers as expected there's been a deluge of information as microsoft took the lid off of its windows 11 announcement and the ensuing uh what's known in the technical world as big oof that followed and this was because of microsoft's cpu and tpm requirements which stirred up no shortage of both confusion and criticism initially microsoft stated that cpu support would begin with intel's eighth generation or eighth series that would be 8700k for example 8600k and also that would begin with amd's ryzen 2000 series ryzen 1000 the initial came out around 2017 or so and uh it's also around when the eighth series came out from intel now this in itself didn't disqualify the older cpus if you look at windows 10 for example there are plenty of cpus that are technically unsupported that support windows 10 or run windows 10 just fine but it was no less disappointing to see microsoft so hyper focused on newer hardware and of course you you think outside the consumer space where people are a little more conscious about what they're spending using you look at the corporate spaces and most of the big companies are going to play exactly by the book and follow the specs that it officially supports so that does start to limit what's usable with windows 11 in those environments now in an effort to smooth over the hardware concerns among others microsoft stated in a blog post that it's working to validate systems using intel's 7th gen cpus which came out about january of the same year as the 8th gen cpus and which are functionally the same as intel's 6th and 4th gen cpus and we don't talk about 5. so they're doing a little bit better they're also adding ryzen 1000 cpu official support but that's where it stops for both of them additionally microsoft's received no shortage of criticism for its laughable pc health check app which offered literally zero insight as to why a machine was or wasn't supported in fact the open source app why not win 11 proved more useful online as such microsoft announced that it has pulled the app from downloads citing the need to address user feedback namely that it wants to make the tool do something it said quote based on the feedback so far we acknowledge that it was not fully prepared to share the level of detail or accuracy you expected from us on why a windows 10 pc doesn't meet upgrade requirements there was also some level of confusion around windows 11 would roll out to be clear windows 11 has a free upgrade path to current and eligible windows 10 users though those roll outs won't happen until some point during the first half of 2022 microsoft will begin rolling out certified win 11 devices in late 2021 they say around october and it noted that the rollout would take place across several months microsoft just released the first preview build of windows 11 to windows insiders where it expects to collect feedback in a number of areas because obviously microsoft noted that during this time the feedback will inform its decision on the minimum hardware requirements going forward up next team group is the first memory vendor that we've seen anyway out the door with available ddr5 modules despite the dearth of cpus and motherboards actually supporting the the new memory the emergent memory team groups elite ddr5 4800 kit has surfaced first and it has listings on both amazon and newegg right now both of the retailers do show this kit out of stock or unavailable so whether you can get it's a different story but team group's elite ddr5 4800 kit is 32 gigabytes 16x2 with two 16 gigabyte udem modules rated at 4800 mega transfers per second as expected the dimms come with ondi ecc and an onboard pmic team group's first ddr5 kit also has timings of cl 40 40 40 77. it's got a cast latency of 40 and an operating voltage of 1.1 volts both retail listings show the kit price at 311 u.s and it looks like they're expected to ship at some point in july the first consumer platform to support ddr5 will be intel's alder lake that platform is currently not a well it's officially announced until did show it in its disaster of a keynote again uh but it's rumored for an ounce late 2021 we don't have much information about it beyond that around the other on the release date and it'll be a while before you can actually really use ddr5 in the consumer market anyway so ddr5 will also take a while to scale beyond the best of ddr4 as a reminder the switchover period from when stuff changes like say from ddr3 to ddr4 that period of time the new memory is not advanced to the point where if you have really high-end existing memories ddr4 uh versus brand new what will eventually be low end ddr5 there's not normally that bigger performance gap there and sometimes it favors the older kit instead it just depends on the architecture of the cpu and the platform you're buying for of course thanks steve up next the usgpu market changing so as a result of china's crackdown on cryptocurrency the first wave of used mining gpus at least on mass has appeared online for the aftermarket and this was reported by pc gamer where screenshots posted to the taiwanese bulletin board ptt was we're showing a growing count of nvidia and amd gpus for sale and the gpus range from nvidia's rtx 30 series and amd's rx 6000 series at the newer side of things two older cars like of course and these polaris based cars that'd be the rx 400 500 series ethereum miners have been especially fond of the polaris based cards in the past so it's not really much of a surprise that they're starting to pop up on ptt according to hkepc the used price for an rtx 3070 is about 3120 hong kong dollars which works out to around 400 usd other cards are priced similarly the rtx 3060 ti for example is around 309 usd the 3060 non-ti is 244 however the cards are also being sold in bulk and they are also used mining carts which means they've likely been in service 24 7. it's the bulk aspect though that'll make it harder to deal with that said this is the latest of several developments that could mean gpu prices and availability are coming back down to earth we recently reported that asrock was optimistic about gpu price and stock uh especially as it expects amd's supply of gpus will improve during the second quarter or at this point rolling to towards third quarter of 2021. it's not just gpu shortages as in the the physical piece of silicon that were a problem though it's things like substrates and components too azeroth commented on that last week we talked about it as well where it said it expects substrate shortages to improve later this year uh now moreover popular coins like ethereum and bitcoin at least recently as of this film changes all the time have dropped in profitability this also affects the uh the dumping of used mining cards on the market and how fast that's happening ethereum's moving eventually in theory to a proof of stake model which moving away from proof of work will also be beneficial to gpu supply so things might be lining up in a way that's beneficial for those of you who still need to get in on a gpu even if it is used and it's just a matter of of well how the cryptocurrencies behave at this point because they seem completely erratic and difficult to predict so who knows we'll see what happens uh micron will sell its 3d cross point fab to texas instruments texas instruments of course also known as thai micron just announced that it would sell off its lehigh utah fab previously when it said it would exit the 3d crosspoint memory business and micron intended to have the fab sold by end of 2021 although at that time it disclosed no details on potential sale options micron recently announced that its 3d crosspoint memory fab will be sold to texas instruments micron states that the economic value of the sale is 1.5 billion dollars with 900 million coming from ti for the purchase of the facilities and the remaining 600 million coming from secondary sales like fab tooling and assets micron's press release didn't state much of what texas instruments plans to do with the fab not really much of a surprise there other than to deploy its own technologies ti will be can't say that without thinking of tie now ti will also attempt to retain all the fab's existing employees it says once the transaction closes it also said in its own press release that it plans to overhaul the fab for 65 nanometer and 45 nanometer production for analog and embedded processors once the purchase closes this will be ti's fourth fab of this type ti currently has dmos 6 rfab 1 and the soon to be completed rfab 2 way for fabrication facilities worthy of note is just how long intel will be able to buy 3d crosspoint memory from micron 4 at this point the wafer agreement between the two parties expires at the end of 2021 and with this sale and microns exit from the business a new agreement will obviously not be possible intel will have to set up its own lines for any products using octane memory in the future if it decides to keep making it although a lot of that starts getting into data center and enterprise markets that we don't cover too closely up next the same model responsible for the resurrection of the 3dfx voodoo 5 6000 that reported on previously has brought back another gpu from the debt these are fun projects they're really cool it's not like you can do a whole lot with it other than think that it's cool but that's nice to have every now and then too so the modder we speak of is anthony i'm not gonna try sorry we'll just we'll put the name on the screen my apologies and i was documented by a youtube video from michael dale which is a name that i can say i think the latest creation is a combination of a 16 megabyte voodoo 3 3500 and a 4 megabyte power vr device that's the pc x2 the design is a custom creation from we're going to go with that normally i use the last name but we're going to go with anthony who has combined both of the aforementioned retro gpus atop a black pcb which looks great by the way and the voodoo 3500 runs at 183 megahertz that's right get in line for this one it's gonna this will be the the gpu of the future voodoo i i think they're going to be the one nvidia is probably not long for this world so the 183 megahertz gpu has access to 16 megabytes of sd ram via eight sdram packages the pcx2 runs at 66 megahertz with four megabytes of sdram the board also features a pass-through port and circuitry to allow for another graphics card to be looped in and the board also features a special jumper for overclocking the pcx2 to 80 megahertz both gpus make use of passive heat sinks although it seems active cooling is recommended depending on the clock speeds the board also uses a standard pci slot rather than a pcie slot and in all seriousness if someone could post a guide on how to say that name i'd really appreciate the guide because it'll probably come up again tried using google i don't trust the results i got but let me know how to say it we'll do that next time uh next time we come across one of these awesome mods by anthony all right samsung and synopsis collaborating on three nanometer gaa process this will be our last story for this week samsung and synopsys announced that through their collaboration samsung has been able to achieve first pass silicon for an soc using samsung's 3gaa process samsung's three gaa process is based on what samsung is calling three nanometers and uses gaa phets or gate all-around transistors these types of transistors are largely the future for scaling silicon and we've discussed them numerous times and news stories recently neither company went into details on the gaa transistors being used samsung at this point has long been working on a proprietary mbc phet which is a multi-bridge channel fed and it's based on nanosheets but like other gaa fats nanosheets will allow for tuning performance or power consumption by augmenting the nanosheet width there's a quote here as well the gaa transistor structure marks a key inflection point in process technology advancement that's critical in maintaining the scaling trajectories needed for the next wave of hyperscale innovation by synopsis in addition to this tape out samsung's customers will have access to synopsis technology files through samsung foundry so presumably samsung foundry clients will also have access to similar designs and synopsis if you don't know is a silicon design company that's focused on design verification ip and software that's it for the news this week thanks as always for watching you can subscribe for more go check out the channel for some of our follow-up dg1 standalone video card coverage uh or you can go to store.camerasnexus.net to grab one of our mouse mats or patreon.com to see that new behind-the-scenes video thanks for watching we'll see you all next time\n"