AMD's Renoir Processor Offers Improved Memory Controller with LPDDR4X Support

AMD has recently released its new processor, Renoir, which features an improved memory controller supporting LPDDR4X X 4266 memory. This is a welcome upgrade over the Raven Ridge and Picasso processors, both of which support DDR4 2400. The use of LPDDR4X X memory should theoretically benefit AMD in a big way, as this type of memory offers faster speeds and lower power consumption compared to traditional DDR4 memory.

This increased memory support will also give AP processors an advantage, as they do not have onboard memory like Intel's CPUs or dedicated video cards. Instead, they rely on system memory, which can be faster and more responsive when equipped with high-quality LPDDR4X X memory. This is especially important for applications that require low-latency data access, such as real-time rendering or physics simulations.

The introduction of broader memory support to AMD's processors is seen as a necessary step to compete with Intel's forthcoming Ice Lake Mobile chips, which will feature DDR4 3200 LPDDR4 3733 memory. The higher memory specifications should give AMD an edge in terms of performance and power efficiency, particularly for applications that rely on system memory.

Intel's Cascade Lake Axe Processor Set to Ship in October

September was a busy month for CPU releases, with several new processors hitting the market. However, next month could be even busier, as Intel is expected to release its Cascade Lake Axe processor, which will feature improved Gen 11 graphics and higher boost clocks compared to its predecessor.

The Cascade Lake Axe processor will ship in October, according to Intel's announcement earlier this year. The company has been working hard to improve the performance and efficiency of its processors, and this new release is seen as a major step forward. The processor will also feature a higher boost clock across all cores, although the exact increase is not yet clear.

Intel's aggressive approach to the CPU market is seen as a response to AMD's growing popularity among PC gamers and content creators. In an interview with Ryan Shrout of PC Perspective, Intel's Andy Nguyen acknowledged that the company is under pressure from competitors like AMD. "We're not taking this sitting down," he said. "We see the competition and we see the landscape as it is. We want to give our customers the best products possible."

AMD's Growing Share in the CPU Market

According to a recent survey conducted by Valve, AMD has seen an increase in its share of the CPU market over the past few months. The survey, which gathered data from over 10 million Steam users, shows that AMD now accounts for 19% of the market share, up from 12% in July.

Intel, on the other hand, has seen a decline in its market share, dropping to 81% according to the survey. However, the GTX 1060, GTX 1050 Ti, and GTX 1050 are still the most popular GPUs among Steam users, although their market share is slipping.

The RT X series continues to gain traction, with the RT X 2070 being the most popular model among Steam users, followed closely by the RT X 2060. The RX 580 is also seeing growth in its market share, up 7% over last month.

ASUS's Triton 500 Laptop with 300Hz Display Expected to Ship in October

Acer has announced a new laptop that will feature a 100Hz display and AMD FX processors. However, the most exciting news is the upcoming release of ASUS's ROG Zephyrus S GX 701, which is expected to ship in October.

The laptop will feature an RT X 2080 GPU, which should provide smooth performance for demanding games and applications. The ROG Zephyrus S GX 701 is also expected to have a high refresh rate display, although the exact value is not yet clear.

ASUS has also released pricing information for its Triton 500 laptop, which will be priced at $2,800 in August. However, the company notes that the price may vary depending on the region and availability.

"WEBVTTKind: captionsLanguage: eneveryone welcome back to another hardware news recap for the week this is a big episode so lots of stuff in here we have some thread Ripper leaks to GN through an exclusive source and these are actually pretty interesting ones Intel and it's marketing need work USB 4.0 specifications are actually published now and then discussion from AMD on BIOS updates to fix clock boosting problems among other things before that this video is brought to you by Thermaltake c 360 DD c hard tubing water cooling kit if you're ready to dip your toes into the water and build your first open-loop cooling system the Thermaltake c 360 DD c hard tubing kit comes with all the components you need the kit includes a three sixty millimeter radiator three 120a RGB fans a copper w-4 a RGB water block for the cpu a pump and res DDC combo and all the fittings needed to build a full CPU open-loop learn more at the link in the description below first stop thread ripper leaks to GN we received information from a source in the industry that actually it was an official AMD internal document that were received from said source and it points toward information for the new thread ripper cpu so thread it for three for the cpu and these names are a bit funny but we receive documentation though on it and basically what we've learned is that there's going to be a split so there's a workstation and an h EDT threaded for line of CP years there's a schism between them and the core account and thread count in these documents were not revealed the documents are intended for people like motherboard manufacturers to decide their thermal design thermal solutions for motherboards things like that so it's information you need for heatsink design for example and what this did say though without core thread count was a couple of other things so it did claim 512 kilobytes of l2 cache per core if this ends up being accurate documentation there was omitted l3 cache information it's had four channels ddr4 for one of the versions which is going to be called the str x4 unit and then the other one the SW r x8 will have eight channels that's the workstation unit the other ones h EDT typical thread ripper used to and these document notes that STR x4 and SW r x8 are single socket solutions client platforms that use surface mount LGA sockets and they support the family 17h models 30 through fh or 3-0 H through f/8 3fh and that's just hex codes for the different CPUs so this seems to include previous generation CPUs but we're not on a percent positive on that right now the document also includes a note that str x4 should include 64 lanes of PCI a Gen 4 with 16 of those being switchable to say that if you want to do that as that's the motherboard maker and then the other one SW r x8 is supposed to have 96 to 128 lanes so it might be variable based on the processor of Gen 4 and that allows for 32 switchable lanes to say that if desired ask for overclocking support STR x4 is marked as yes and SW r x8 is marked as no for overclocking it does seem that SW r x8 is basically a budget version of epic or a poor man's epic if you will so the document also notes thermal design requirements it's there's a lot of information you're not going to go through all of it today but what we wanted to report on first was that it states that the CPUs will have to infrastructure groups there's group and there's Group B group a is for the h EDT processors and that assumes a 32 degrees Celsius ambient temperature in determining the thermal solution Group B is for workstation processors and that assumes a higher ambient temperature of 42 degrees Celsius group a has a TDP presently noted as 280 watt whereas a group B does as well whereas T case is marked as 60 degrees Celsius maximum for group a and a T control maximum of 100 degrees Celsius and there's some information on SW rx-8 as well we'll put that chart on the screen or something so we also received voltage requirements other information on this but we need we require more time to crawl through the rest of the documentation it's fairly extensive and see if we can piece together the more interesting or useful information before launch we'll put that in the next news video but this will get you started separately if you are curious about really states we don't have that information directly from anybody hardware info did though add a patch note that said TR 3,000 support a preliminary support has been added so obviously they're ramping up for it on the software side until recently reembarked on its realistic workload benchmarks at Aoife or IFA so Intel's marketing needs work just the name of this segment until called out and e4 leaning so heavily on Cinebench and testing we do agree that Andy does too much Cinebench and that's not the best place to be but Intel's approach to this isn't really much better so roughly in the July timeframe previously Intel noted that its core i9 processors will perform better and quote Windows desktop applications and that web performance is also better on Intel whatever that means and also triple-a PC gaming more recently the EFI discussion as in like in the past week that talks about notebooks and uses two and one utilization of rendering applications as an indicator of the relevance of said applications it ends up downplaying the importance of rendering applications so a few things here first of all we agree and we have shown in our CP reviews 3900 x for you whatever that Intel performs better in gaming applications when strictly comparing skewed ask you all the value proposition is not eternally in Intel's favorite kind of varies until we think has forfeited the i-5 segment to our fives although the 9600 K is technically a leader in Raw FPS the gap is small enough and the deficit elsewhere is large enough that we no longer recommend i5 CPU is and instead point toward r5 3600 CPUs that stated Intel's 9700 K is better positioned against the 3700 K in gaming while having comparable price and the 9900 K obviously is still the chart leader and raw game in perforins al biet it is expensive now if you're someone who says I only play games I want high refresh rate I want to buy the highest end GPU possible what should I buy and and and also I have a budget that doesn't really matter too much which cpu I buy then yeah we'll say 9900 Kay is the chart-topper for that use case but once you start looking outside of that use case it gets really difficult especially for things like production tasks for Intel to continue to prove its value in the r5 and i-5 segment AMD is the clear choice the r5 3600 is extremely good so let's go over this a bit 4-1 Intel is downplaying production performance overall where Andy does well this should surprise nobody and he does the same thing with regard to Intel's gaming performance basically says why do you need that many frames who cares it's good enough we're within X % so they both do this to each other but Intel does this primarily by beating the same dead horse that Cinebench isn't representative of all use and we agree that is not however there's an important point here Cinebench although we do not use it for reviews and we don't really like it for a lot of reasons Cinebench is not that different from what you might what Intel might call real applications blender is a pretty damn real application it's a rendering application Cinebench is representative of cinema 4d which is used by people blender is a growing pretty wide use user base and we use it internally for our own rendering so we're really comfortable with using it for benchmarking because we know exactly what's needed and we render our own scenes that we've built so it's like real use cases and it's pretty hard to argue in that scenario that is not representative of real use because literally real use like we we actually used blender to do those like the intro animation of the videos so Cinebench isn't that different from blender in that regard there both rendering applications so while Intel is downplaying Cinebench it is by collateral also downplaying other tile based rendering applications and that is a bad thing to do because they are very popular even if not everybody uses them like I don't know say for example photo shops pretty widespread from here is pretty widespread rendering applications like cycles renderer in I should be saying 3d applications like blender not rendering applications but 3d applications very widespread of course especially look at the games and the movie industries so all of that stead it's AMD still went even blender so else in Duncan downplay Cinebench and these still winning in real applications using the they're the real terminology that until you dislike blender and it's not that dissimilar so that's a ami does well with Cinebench for a few reasons one of them is it can fit everything in cash that it needs for the render and so it doesn't really need to reach outside to grab any data this benefits AMD but tile based rendering is also used elsewhere and if you start looking at applications like Cinebench vs blender the kind of irony amusing irony here is that while Intel is saying Cinebench is bad or perhaps not that exact phrasing if they I'm sure I'm sure if I get contacted by Intel they'll say we didn't say it's bad Intel is saying it is not realistic let's go with that well then tell saying it's not realistic the the I I think amusing part is Intel will relatively do better in Cinebench than it will in other 3d rendering applications because Cinebench by the bench indicator in the name is meant to be a benchmark and it is therefore pretty short and pretty easy to run and because it's short Intel with its turbo boosting duration limits enabled properly on motherboards Intel will do better in Cinebench because if you run it once especially r15 you run it once you don't run it in rapid succession what you end up with is the CPUs boost higher to hit that maximum power state that it's allowed to meet for however many seconds hundred seconds whatever it is twenty seconds for the first 100 for the second one - something like that it's been on the board a bit Intel will reach that boost for the duration of the Senate managed benchmark and therefore look relatively better versus AMD when compared to something like blender where you might be rendering for the period of time to turn by the benchmark and in our case because we're using a real slide that we made a real frame from an animation it takes about 30 minutes on a mid-range CPU maybe 10 on a really on CPU so no matter what your well beyond that turbo-boost duration limit and therefore until will exit the turbo for all for that period after the periods expired it exits it clocks down runs at the 90 ish watt load for the rest of the benchmark and will look relatively worse versus AMD so Intel could actually I mean like you know you're not winning and either of them but relatively it should actually look better in Cinebench on average anyway all that stated the 3900 X also does better in our premier benchmarks than the comparable Intel CPUs which actually surprised us initially didn't used to be that way so there's a massive blow to a segment where Intel was previously advantaged Intel still has a very strong advantage and Photoshop to be fair but that's the only one really where we're seeing frequency so heavily benefited out of the benchmarks we're on so friend of the site Rob Williams of tech gauge comm recently wrote an article talking about this more about Intel's new marketing and he called out how Intel has begun contradicting its previous marketing which he is absolutely accurate on so previously Intel worked with Mac's on and worked on sand events used Cinebench to demonstrate its performance with 18 core processors for example and now that AMD is using the same trick against Intel Intel has decided that the previous marketing is no longer useful so that's the story on that some some bits of irony in there but mostly just stupid big companies fighting each other I think Roman actually said it the best the Roman okay there Bauer on YouTube I believe his phrase was that when you put the two slides next to each other and tell any of these slides next to each other it's one high school fifteen-year-old girl fighting another high school fifteen-year-old girl and that was how he perceives the battle between the two I would say Roman as usual your German precision is dead on USB for specifications published so USB implementers forum or USB eye F has officially announced the new USB 4 specification and note that there is a difference in the spacing between where the forgoes versus USB because you SPE groov wants to continually change the name so that no one understands anything but USB for with that space no longer between USB and the number critical change you miss before offers twice the bandwidth of USB 3.2 it's a theoretical 40 gigabits per second of throughput and that's assuming the use of the certified cables mind you u.s. pif announced us before earlier this year as well as revealing some confusing new branding for the existing generations of USB in addition to the increased bandwidth USB for will be backwards compatible with previous USB specifications just like the last ones have been there's also a universal Thunderbolt 3 compatibility and Intel offered up the Thunderbolt 3 spec to the usb if' meaning that any vendor wanting to offer a Thunderbolt 3 compatible device no longer needs to license to use the technology however the catch is that Thunderbolt 3 isn't technically required for USB 4 and so implementation is optional additionally any device that is Thunderbolt 3 compatible will Note will need to undergo validation which is not free and it remains to be seen how well adopted Thunderbolt 3 will become obviously a USB 4 will also allow or offer better allocation between data and video bandwidth and it will retain the type-c connector however USB if' will announce new specifications for the type-c connector in the future to handle the new USB 4 interface power and data delivery stuff like that all right next up speaking of their Bauer earlier de Brouwer survey and the wake of Rison 3000 to issues with reaching the advertised boost clocks there Bauer commissioned a survey among the user base to assess how pervasive the issue was the results were seemingly worse than expected and this all came about because in speaking with romanette LTX earlier in person maybe two months ago this came about because when we were all testing rise in 3000 behind the scenes before embargo left Roy and I were talking we were seen boost issues we published probably like a week's worth of content on frequency issues with Rison 3000 Roman was having issues with it as well that were different than ours and Andy talking to Roman was more or less saying no one else has seen this and us talking to Andy at the same time without AMD having knowledge of these of this like triangle existing we were saying hey we're having frequency issues so Rowan was getting word that no no one else's you're doing it wrong or at least the implication thereof and that wasn't true so he in deciding to be absolute determine how bad the issue is collected something like three thousand results and we recommend watching his video it's on their Bower on YouTube the channel will link in our show notes below though so there's been an abundance of confusion to recap this over rise in 3,000 views clock behavior enough to warrant Andy to tail back and update its rise in product pages something we talked about last week confirming that only one core is awarded the highest boost clock that should surprise no one but it's deeper than that and Andy didn't really go much deeper than that at his product updates so still many users haven't achieved the specified boost frequencies on any of the cores at all this is something we showed in our initial reviews for the 3900 X certainly and some of the others 3,600 we actually were reaching the boost spec on but that was the only one there Bower survey shows that we weren't alone and being having difficulty reaching these specs and he pulled 2700 participants who ran the single threaded benchmark on Cinebench r15 the maximum clock speed was recorded with hardware info for that and these are our five thirty six hundred seems to fare the best matching our own data that we saw same thing and it had roughly half users reporting boost clocks as advertised they're in worse however was the our 939 hundred X with only 5.6 percent of users seeing the correct maximum boost clocks and it seemed as many boost clocks were anywhere between 25 to 100 megahertz shy of the correct clock speeds and somewhere further off than that so we're only going to show a little bit was like maybe one of his charts in the video he has a lot of data please go watch it it's well worth it he's done a lot of work on it and deserves the viewership so again that'll be linked below if you want to see it and the pushing BIOS updates to fix boost clock problems this is certainly completely unrelated to the previous news topic and has nothing to do with it or what Romans done so with the aforementioned news this leads to an DS official response that rise in 3,000 boost clock issues are being addressed probably andis been looking at this for a little while but the recent public pressure from media hasn't made it easy so Andy took to Twitter to both acknowledge and update the community regarding the problem Twitter of course being the official source of news for anything and said quote Andy is pleased with the strong momentum of 3rd gen AMD risin processors and the PC enthusiasts and gaming communities we closely monitor feedback on our products and understand that some third generation users are reporting some are reporting PC clock speeds below the expected processor frequency while processor boost frequency is dependent on many variables including workload system design cooling solution we have closely reviewed the feedback from our customers and have identified an issue in our firmware that reduces the boost frequencies in some situations continue the quote saying we are in the process of preparing a BIOS update for our motherboard partners that addresses that issue and includes additional boost performance optimizations will provide an update on September 10th to the community regarding availability of the BIOS we've already GM's already been in contact with motherboard manufacturers we have the early versions of that firmware coming in as soon as they arrive and a couple things here this is a very pyaari statement in that Andy is doing the usual PR thing of saying starting off by saying everything's amazing thank you for the extremely strong momentum there's a small problem a few people about 95% aren't reaching the advertised boost clock so for those few people with 3,900 X 95% of you we will fix that problem so anyway well test it as soon as we get it and the renoir AP used to come with improved memory support this is a quick one a pair of Linux patches have ostensibly confirmed that AMD's upcoming renoir ApS will come equipped with and improved memory controller supporting lpddr4 X 4266 memory this would be a welcome upgrade over the Raven Ridge and Picasso processors and both of those support ddr4 2400 and an official capacity Renoir looks to be an these first chip supporting lpddr4 X memory and we could make the argument that AMD needed to introduce broader memory support to these types of products to compete with Intel's forthcoming ice like mobile chips which support ddr4 3,200 lpddr4 3733 for example and will feature improved gen 11 graphics the higher memory specifications should theoretically benefit AMD in a big way because with AP use you don't have onboard memory with like a mother or a video card rather so because you don't have a GPU that can just talk to its own dedicated memory that's faster and physically very close having faster system memory benefits ap is a lot because they do they do lean on those and then a couple quick ones here so nine hundred KS and cascade Lake axe shipping in October as mentioned previously October could be a crowded month for CPUs high-performance CPUs and he's got its new stuff coming out in September in theory and for Intel Ryan sharp previously a PC perspective took to the stage at Intel's even presentation the same one discussed a second ago to announce that October ship date for the cascade Lake axe h EDT parts and the binned nine nine hundred KS CPUs is is coming up fast so Shrout was scarce stuff with details but noted that the nine hundred KS will come with a five gigahertz boost clock across all cores we talked about this previously prior on Computex that's not too distant from what it does already but if it's been then the real hope we hope is that we can get it in a stream to smell into and push it higher and actually speaking about to backtrack in a second AMD statement where they said that boosts clocks depend on thermal solution i should note that we didn't reach the maximum advertised boost until we hit minus eighty degrees celsius so if you are if if your thermal solution you know is anywhere north of that then i guess you need a new one anyway back to intel AMD and it's rising three thousand chips were pointed that in the intel presentation for the same issues discussed earlier shroud also noted that the increased pressure andy has placed on intel is noted and said quote the point says we're not taking this sitting down we see the competition and we see the landscape as it is we're adjusting because we take these customers very seriously and we want to give them the best products possible which is actually kind of a big thing from intel which has historically more less ignored that anthe has existed Intel's a big enough company that I can take the approach of just doing what it does and when you say what about those new AMD FX processors they say the new huh what is that ACS and Acer gaming laptops with a hundred Hertz display is coming out and defending why you need 300 Hertz displays they didn't share any configuration details or at least a sir didn't but we can assume a sir will ship the Triton 500 with an RT X 2080 as well like the ASIS one we can speculate that both companies are sourcing the panel from the same place asus is expected to ship the rog zephyrus s GX 701 and october price TBA with a survey in a december launch for its triton 500 priced at $2,800 if you're interested in 300 displays on the laptop and then august steam hardware serving last one valve released its steam hardware survey for august and the biggest highlight is the trend of AMD increase in its cpu share among survey participants a trend that's been ongoing since July and II now accounts for 19% of CB share among Steam users who participated in the survey while intel has dropped to 81 percent the GTX 1060 1050 Ti and 1050 are still the most popular GPUs by way of the survey albeit their grasp is slipping meanwhile the RT X series continues to make steady progress and NVIDIA super refresh has likely contributed here the RT X 2070 is still the most popular RT x model among Steam users that is a point one nine percent over the last month the RT X 2060 comes in as a close second up 0.24 percent over the last month and the RX 580 is Andy's most popular card currently still seen growth with a point zero seven percent increase in share over last month that's it for the news video I said this was a pretty packed one you want to helps out directly with this type of stuff you can go to store documents nexus net to become one of our shirts one of our mod mass toolkits or something else and you can also go to patreon.com/scishow sack says thank you for watching we'll see you all next timeeveryone welcome back to another hardware news recap for the week this is a big episode so lots of stuff in here we have some thread Ripper leaks to GN through an exclusive source and these are actually pretty interesting ones Intel and it's marketing need work USB 4.0 specifications are actually published now and then discussion from AMD on BIOS updates to fix clock boosting problems among other things before that this video is brought to you by Thermaltake c 360 DD c hard tubing water cooling kit if you're ready to dip your toes into the water and build your first open-loop cooling system the Thermaltake c 360 DD c hard tubing kit comes with all the components you need the kit includes a three sixty millimeter radiator three 120a RGB fans a copper w-4 a RGB water block for the cpu a pump and res DDC combo and all the fittings needed to build a full CPU open-loop learn more at the link in the description below first stop thread ripper leaks to GN we received information from a source in the industry that actually it was an official AMD internal document that were received from said source and it points toward information for the new thread ripper cpu so thread it for three for the cpu and these names are a bit funny but we receive documentation though on it and basically what we've learned is that there's going to be a split so there's a workstation and an h EDT threaded for line of CP years there's a schism between them and the core account and thread count in these documents were not revealed the documents are intended for people like motherboard manufacturers to decide their thermal design thermal solutions for motherboards things like that so it's information you need for heatsink design for example and what this did say though without core thread count was a couple of other things so it did claim 512 kilobytes of l2 cache per core if this ends up being accurate documentation there was omitted l3 cache information it's had four channels ddr4 for one of the versions which is going to be called the str x4 unit and then the other one the SW r x8 will have eight channels that's the workstation unit the other ones h EDT typical thread ripper used to and these document notes that STR x4 and SW r x8 are single socket solutions client platforms that use surface mount LGA sockets and they support the family 17h models 30 through fh or 3-0 H through f/8 3fh and that's just hex codes for the different CPUs so this seems to include previous generation CPUs but we're not on a percent positive on that right now the document also includes a note that str x4 should include 64 lanes of PCI a Gen 4 with 16 of those being switchable to say that if you want to do that as that's the motherboard maker and then the other one SW r x8 is supposed to have 96 to 128 lanes so it might be variable based on the processor of Gen 4 and that allows for 32 switchable lanes to say that if desired ask for overclocking support STR x4 is marked as yes and SW r x8 is marked as no for overclocking it does seem that SW r x8 is basically a budget version of epic or a poor man's epic if you will so the document also notes thermal design requirements it's there's a lot of information you're not going to go through all of it today but what we wanted to report on first was that it states that the CPUs will have to infrastructure groups there's group and there's Group B group a is for the h EDT processors and that assumes a 32 degrees Celsius ambient temperature in determining the thermal solution Group B is for workstation processors and that assumes a higher ambient temperature of 42 degrees Celsius group a has a TDP presently noted as 280 watt whereas a group B does as well whereas T case is marked as 60 degrees Celsius maximum for group a and a T control maximum of 100 degrees Celsius and there's some information on SW rx-8 as well we'll put that chart on the screen or something so we also received voltage requirements other information on this but we need we require more time to crawl through the rest of the documentation it's fairly extensive and see if we can piece together the more interesting or useful information before launch we'll put that in the next news video but this will get you started separately if you are curious about really states we don't have that information directly from anybody hardware info did though add a patch note that said TR 3,000 support a preliminary support has been added so obviously they're ramping up for it on the software side until recently reembarked on its realistic workload benchmarks at Aoife or IFA so Intel's marketing needs work just the name of this segment until called out and e4 leaning so heavily on Cinebench and testing we do agree that Andy does too much Cinebench and that's not the best place to be but Intel's approach to this isn't really much better so roughly in the July timeframe previously Intel noted that its core i9 processors will perform better and quote Windows desktop applications and that web performance is also better on Intel whatever that means and also triple-a PC gaming more recently the EFI discussion as in like in the past week that talks about notebooks and uses two and one utilization of rendering applications as an indicator of the relevance of said applications it ends up downplaying the importance of rendering applications so a few things here first of all we agree and we have shown in our CP reviews 3900 x for you whatever that Intel performs better in gaming applications when strictly comparing skewed ask you all the value proposition is not eternally in Intel's favorite kind of varies until we think has forfeited the i-5 segment to our fives although the 9600 K is technically a leader in Raw FPS the gap is small enough and the deficit elsewhere is large enough that we no longer recommend i5 CPU is and instead point toward r5 3600 CPUs that stated Intel's 9700 K is better positioned against the 3700 K in gaming while having comparable price and the 9900 K obviously is still the chart leader and raw game in perforins al biet it is expensive now if you're someone who says I only play games I want high refresh rate I want to buy the highest end GPU possible what should I buy and and and also I have a budget that doesn't really matter too much which cpu I buy then yeah we'll say 9900 Kay is the chart-topper for that use case but once you start looking outside of that use case it gets really difficult especially for things like production tasks for Intel to continue to prove its value in the r5 and i-5 segment AMD is the clear choice the r5 3600 is extremely good so let's go over this a bit 4-1 Intel is downplaying production performance overall where Andy does well this should surprise nobody and he does the same thing with regard to Intel's gaming performance basically says why do you need that many frames who cares it's good enough we're within X % so they both do this to each other but Intel does this primarily by beating the same dead horse that Cinebench isn't representative of all use and we agree that is not however there's an important point here Cinebench although we do not use it for reviews and we don't really like it for a lot of reasons Cinebench is not that different from what you might what Intel might call real applications blender is a pretty damn real application it's a rendering application Cinebench is representative of cinema 4d which is used by people blender is a growing pretty wide use user base and we use it internally for our own rendering so we're really comfortable with using it for benchmarking because we know exactly what's needed and we render our own scenes that we've built so it's like real use cases and it's pretty hard to argue in that scenario that is not representative of real use because literally real use like we we actually used blender to do those like the intro animation of the videos so Cinebench isn't that different from blender in that regard there both rendering applications so while Intel is downplaying Cinebench it is by collateral also downplaying other tile based rendering applications and that is a bad thing to do because they are very popular even if not everybody uses them like I don't know say for example photo shops pretty widespread from here is pretty widespread rendering applications like cycles renderer in I should be saying 3d applications like blender not rendering applications but 3d applications very widespread of course especially look at the games and the movie industries so all of that stead it's AMD still went even blender so else in Duncan downplay Cinebench and these still winning in real applications using the they're the real terminology that until you dislike blender and it's not that dissimilar so that's a ami does well with Cinebench for a few reasons one of them is it can fit everything in cash that it needs for the render and so it doesn't really need to reach outside to grab any data this benefits AMD but tile based rendering is also used elsewhere and if you start looking at applications like Cinebench vs blender the kind of irony amusing irony here is that while Intel is saying Cinebench is bad or perhaps not that exact phrasing if they I'm sure I'm sure if I get contacted by Intel they'll say we didn't say it's bad Intel is saying it is not realistic let's go with that well then tell saying it's not realistic the the I I think amusing part is Intel will relatively do better in Cinebench than it will in other 3d rendering applications because Cinebench by the bench indicator in the name is meant to be a benchmark and it is therefore pretty short and pretty easy to run and because it's short Intel with its turbo boosting duration limits enabled properly on motherboards Intel will do better in Cinebench because if you run it once especially r15 you run it once you don't run it in rapid succession what you end up with is the CPUs boost higher to hit that maximum power state that it's allowed to meet for however many seconds hundred seconds whatever it is twenty seconds for the first 100 for the second one - something like that it's been on the board a bit Intel will reach that boost for the duration of the Senate managed benchmark and therefore look relatively better versus AMD when compared to something like blender where you might be rendering for the period of time to turn by the benchmark and in our case because we're using a real slide that we made a real frame from an animation it takes about 30 minutes on a mid-range CPU maybe 10 on a really on CPU so no matter what your well beyond that turbo-boost duration limit and therefore until will exit the turbo for all for that period after the periods expired it exits it clocks down runs at the 90 ish watt load for the rest of the benchmark and will look relatively worse versus AMD so Intel could actually I mean like you know you're not winning and either of them but relatively it should actually look better in Cinebench on average anyway all that stated the 3900 X also does better in our premier benchmarks than the comparable Intel CPUs which actually surprised us initially didn't used to be that way so there's a massive blow to a segment where Intel was previously advantaged Intel still has a very strong advantage and Photoshop to be fair but that's the only one really where we're seeing frequency so heavily benefited out of the benchmarks we're on so friend of the site Rob Williams of tech gauge comm recently wrote an article talking about this more about Intel's new marketing and he called out how Intel has begun contradicting its previous marketing which he is absolutely accurate on so previously Intel worked with Mac's on and worked on sand events used Cinebench to demonstrate its performance with 18 core processors for example and now that AMD is using the same trick against Intel Intel has decided that the previous marketing is no longer useful so that's the story on that some some bits of irony in there but mostly just stupid big companies fighting each other I think Roman actually said it the best the Roman okay there Bauer on YouTube I believe his phrase was that when you put the two slides next to each other and tell any of these slides next to each other it's one high school fifteen-year-old girl fighting another high school fifteen-year-old girl and that was how he perceives the battle between the two I would say Roman as usual your German precision is dead on USB for specifications published so USB implementers forum or USB eye F has officially announced the new USB 4 specification and note that there is a difference in the spacing between where the forgoes versus USB because you SPE groov wants to continually change the name so that no one understands anything but USB for with that space no longer between USB and the number critical change you miss before offers twice the bandwidth of USB 3.2 it's a theoretical 40 gigabits per second of throughput and that's assuming the use of the certified cables mind you u.s. pif announced us before earlier this year as well as revealing some confusing new branding for the existing generations of USB in addition to the increased bandwidth USB for will be backwards compatible with previous USB specifications just like the last ones have been there's also a universal Thunderbolt 3 compatibility and Intel offered up the Thunderbolt 3 spec to the usb if' meaning that any vendor wanting to offer a Thunderbolt 3 compatible device no longer needs to license to use the technology however the catch is that Thunderbolt 3 isn't technically required for USB 4 and so implementation is optional additionally any device that is Thunderbolt 3 compatible will Note will need to undergo validation which is not free and it remains to be seen how well adopted Thunderbolt 3 will become obviously a USB 4 will also allow or offer better allocation between data and video bandwidth and it will retain the type-c connector however USB if' will announce new specifications for the type-c connector in the future to handle the new USB 4 interface power and data delivery stuff like that all right next up speaking of their Bauer earlier de Brouwer survey and the wake of Rison 3000 to issues with reaching the advertised boost clocks there Bauer commissioned a survey among the user base to assess how pervasive the issue was the results were seemingly worse than expected and this all came about because in speaking with romanette LTX earlier in person maybe two months ago this came about because when we were all testing rise in 3000 behind the scenes before embargo left Roy and I were talking we were seen boost issues we published probably like a week's worth of content on frequency issues with Rison 3000 Roman was having issues with it as well that were different than ours and Andy talking to Roman was more or less saying no one else has seen this and us talking to Andy at the same time without AMD having knowledge of these of this like triangle existing we were saying hey we're having frequency issues so Rowan was getting word that no no one else's you're doing it wrong or at least the implication thereof and that wasn't true so he in deciding to be absolute determine how bad the issue is collected something like three thousand results and we recommend watching his video it's on their Bower on YouTube the channel will link in our show notes below though so there's been an abundance of confusion to recap this over rise in 3,000 views clock behavior enough to warrant Andy to tail back and update its rise in product pages something we talked about last week confirming that only one core is awarded the highest boost clock that should surprise no one but it's deeper than that and Andy didn't really go much deeper than that at his product updates so still many users haven't achieved the specified boost frequencies on any of the cores at all this is something we showed in our initial reviews for the 3900 X certainly and some of the others 3,600 we actually were reaching the boost spec on but that was the only one there Bower survey shows that we weren't alone and being having difficulty reaching these specs and he pulled 2700 participants who ran the single threaded benchmark on Cinebench r15 the maximum clock speed was recorded with hardware info for that and these are our five thirty six hundred seems to fare the best matching our own data that we saw same thing and it had roughly half users reporting boost clocks as advertised they're in worse however was the our 939 hundred X with only 5.6 percent of users seeing the correct maximum boost clocks and it seemed as many boost clocks were anywhere between 25 to 100 megahertz shy of the correct clock speeds and somewhere further off than that so we're only going to show a little bit was like maybe one of his charts in the video he has a lot of data please go watch it it's well worth it he's done a lot of work on it and deserves the viewership so again that'll be linked below if you want to see it and the pushing BIOS updates to fix boost clock problems this is certainly completely unrelated to the previous news topic and has nothing to do with it or what Romans done so with the aforementioned news this leads to an DS official response that rise in 3,000 boost clock issues are being addressed probably andis been looking at this for a little while but the recent public pressure from media hasn't made it easy so Andy took to Twitter to both acknowledge and update the community regarding the problem Twitter of course being the official source of news for anything and said quote Andy is pleased with the strong momentum of 3rd gen AMD risin processors and the PC enthusiasts and gaming communities we closely monitor feedback on our products and understand that some third generation users are reporting some are reporting PC clock speeds below the expected processor frequency while processor boost frequency is dependent on many variables including workload system design cooling solution we have closely reviewed the feedback from our customers and have identified an issue in our firmware that reduces the boost frequencies in some situations continue the quote saying we are in the process of preparing a BIOS update for our motherboard partners that addresses that issue and includes additional boost performance optimizations will provide an update on September 10th to the community regarding availability of the BIOS we've already GM's already been in contact with motherboard manufacturers we have the early versions of that firmware coming in as soon as they arrive and a couple things here this is a very pyaari statement in that Andy is doing the usual PR thing of saying starting off by saying everything's amazing thank you for the extremely strong momentum there's a small problem a few people about 95% aren't reaching the advertised boost clock so for those few people with 3,900 X 95% of you we will fix that problem so anyway well test it as soon as we get it and the renoir AP used to come with improved memory support this is a quick one a pair of Linux patches have ostensibly confirmed that AMD's upcoming renoir ApS will come equipped with and improved memory controller supporting lpddr4 X 4266 memory this would be a welcome upgrade over the Raven Ridge and Picasso processors and both of those support ddr4 2400 and an official capacity Renoir looks to be an these first chip supporting lpddr4 X memory and we could make the argument that AMD needed to introduce broader memory support to these types of products to compete with Intel's forthcoming ice like mobile chips which support ddr4 3,200 lpddr4 3733 for example and will feature improved gen 11 graphics the higher memory specifications should theoretically benefit AMD in a big way because with AP use you don't have onboard memory with like a mother or a video card rather so because you don't have a GPU that can just talk to its own dedicated memory that's faster and physically very close having faster system memory benefits ap is a lot because they do they do lean on those and then a couple quick ones here so nine hundred KS and cascade Lake axe shipping in October as mentioned previously October could be a crowded month for CPUs high-performance CPUs and he's got its new stuff coming out in September in theory and for Intel Ryan sharp previously a PC perspective took to the stage at Intel's even presentation the same one discussed a second ago to announce that October ship date for the cascade Lake axe h EDT parts and the binned nine nine hundred KS CPUs is is coming up fast so Shrout was scarce stuff with details but noted that the nine hundred KS will come with a five gigahertz boost clock across all cores we talked about this previously prior on Computex that's not too distant from what it does already but if it's been then the real hope we hope is that we can get it in a stream to smell into and push it higher and actually speaking about to backtrack in a second AMD statement where they said that boosts clocks depend on thermal solution i should note that we didn't reach the maximum advertised boost until we hit minus eighty degrees celsius so if you are if if your thermal solution you know is anywhere north of that then i guess you need a new one anyway back to intel AMD and it's rising three thousand chips were pointed that in the intel presentation for the same issues discussed earlier shroud also noted that the increased pressure andy has placed on intel is noted and said quote the point says we're not taking this sitting down we see the competition and we see the landscape as it is we're adjusting because we take these customers very seriously and we want to give them the best products possible which is actually kind of a big thing from intel which has historically more less ignored that anthe has existed Intel's a big enough company that I can take the approach of just doing what it does and when you say what about those new AMD FX processors they say the new huh what is that ACS and Acer gaming laptops with a hundred Hertz display is coming out and defending why you need 300 Hertz displays they didn't share any configuration details or at least a sir didn't but we can assume a sir will ship the Triton 500 with an RT X 2080 as well like the ASIS one we can speculate that both companies are sourcing the panel from the same place asus is expected to ship the rog zephyrus s GX 701 and october price TBA with a survey in a december launch for its triton 500 priced at $2,800 if you're interested in 300 displays on the laptop and then august steam hardware serving last one valve released its steam hardware survey for august and the biggest highlight is the trend of AMD increase in its cpu share among survey participants a trend that's been ongoing since July and II now accounts for 19% of CB share among Steam users who participated in the survey while intel has dropped to 81 percent the GTX 1060 1050 Ti and 1050 are still the most popular GPUs by way of the survey albeit their grasp is slipping meanwhile the RT X series continues to make steady progress and NVIDIA super refresh has likely contributed here the RT X 2070 is still the most popular RT x model among Steam users that is a point one nine percent over the last month the RT X 2060 comes in as a close second up 0.24 percent over the last month and the RX 580 is Andy's most popular card currently still seen growth with a point zero seven percent increase in share over last month that's it for the news video I said this was a pretty packed one you want to helps out directly with this type of stuff you can go to store documents nexus net to become one of our shirts one of our mod mass toolkits or something else and you can also go to patreon.com/scishow sack says thank you for watching we'll see you all next time\n"