AMD's DNA Architecture: A New Era for Graphics and Beyond

AMD has been working on its DNA architecture for several years now, and recently, the company has released a detailed white paper that outlines its plans for this new technology. The white paper, which can be found on AMD's website, provides a comprehensive overview of the DNA architecture and its potential applications in various fields.

The DNA architecture is designed to provide improved performance and power efficiency compared to traditional graphics processing units (GPUs). According to the white paper, the DNA family will roll out across nearly every part of the industry, including smartphones, tablets, gaming consoles, cloud gaming services, and a full spectrum of gaming GPUs from low-cost to high-performance.

One of the key benefits of the DNA architecture is its ability to virtualize the GPU, allowing it to be shared between different operating systems. This technology is particularly important for cloud gaming services, which require secure and efficient processing to deliver smooth gameplay. The white paper also mentions that AMD's partnerships with companies like Samsung will bring the DNA-powered graphics to mobile devices in the near future.

The DNA architecture has already been applied to some of AMD's current products, including its Radeon Navi GPUs. However, the company is not stopping there and plans to expand its use of DNA technology to other areas, such as cloud gaming and next-generation consoles. In fact, AMD has partnered with Sony and Microsoft to develop custom solutions for their upcoming consoles, which will take advantage of the DNA architecture's power and efficiency.

The white paper also touches on the topic of cloud gaming and how the DNA architecture can be used to improve performance in this field. The hypervisor agent, a key component of the DNA architecture, enables the GPU to be virtualized and shared between different operating systems. This technology is crucial for cloud gaming services, which require secure and efficient processing to deliver smooth gameplay.

In addition to its applications in graphics and cloud gaming, the DNA architecture has potential uses beyond these areas as well. The white paper mentions that the DNA family will encompass GPUs from low-cost to high-performance, making it a versatile technology with wide-ranging applications.

Intel i5 9600K Price Drop

Recently, Intel has dropped the price of its Core i5-9600K processor, which was previously recommended but not actively recommended due to its lack of Hyper-Threading and other features. The new price point for the processor is $220 on Amazon, which is $20 lower than the previous price drop.

The i5-9600K has six cores and six threads, making it a decent option for gaming and content creation workloads. However, its performance in certain areas is not as strong as other processors with Hyper-Threading, such as the Core i7-9700K. Nevertheless, the processor's overclocking capabilities make it an attractive option for those who want to push their system to its limits.

AMD vs Intel: Which Processor Reigns Supreme?

In our review of the AMD Ryzen 5 3600 and Intel Core i5-9600K, we discussed the pros and cons of each processor. While the i5-9600K has some advantages, such as its higher clock speeds and Hyper-Threading, it also lacks certain features that make it less desirable than other processors.

In contrast, the Ryzen 5 3600 offers a more balanced set of features at a lower price point. Its six cores and twelve threads provide excellent performance for gaming and content creation workloads, making it an attractive option for those who want to build a powerful system without breaking the bank.

Overall, the choice between AMD and Intel depends on individual needs and preferences. While both processors have their strengths and weaknesses, the Ryzen 5 3600 offers a more comprehensive set of features at a lower price point, making it a popular choice among gamers and content creators.

The Future of Gaming: What's Next for AMD?

AMD has been working on its DNA architecture for several years now, and recent developments suggest that this technology will play a major role in the future of gaming. The company's plans to bring the DNA-powered graphics to mobile devices, next-generation consoles, and cloud gaming services are significant milestones in the development of this technology.

As AMD continues to refine and expand its DNA architecture, we can expect to see improved performance and power efficiency across various applications. The company's partnerships with companies like Samsung and Microsoft will also help to bring the DNA-powered graphics to a wider audience.

In conclusion, AMD's DNA architecture is an exciting new development that has the potential to revolutionize the way we experience graphics and gaming. With its focus on power efficiency, virtualization, and versatility, this technology has wide-ranging applications beyond just graphics processing. As AMD continues to develop and refine the DNA architecture, we can expect to see significant improvements in performance and efficiency across various industries.

"WEBVTTKind: captionsLanguage: eneveryone welcome back to another hardware news recap for the week we have some information on NVIDIA thermal monitoring Intel comet-like CPU details Nvidia adding integer scaling with the new drivers and then a couple of AMD stories like the rDNA white paper which is worth checking out and Lisa su commenting on crossfire support before that this video is brought to you by audible audible has a massive audio book library including content that talks computers and games audible has an entire series from the official Computer History Museum which we've actually toured in the past and can support as a leader in computer education audible also hosts the ultimate history of video games something I read back when researching GM content and can highly recommend for gaming and hardware enthusiasts audibles 30-day free trial can be unlocked at audible.com slash gamers Nexus or you can text gamers Nexus one word to 500 500 where you'll get a free audiobook and two audible originals or click the link in the description below so first one's really easy we talked about how AMD does junction temperature reading with the 5700 XT about a week ago and none of that was new that happened in February with the Radeon seven card launch but it was worth recapping because a lot of people with 500 XTC react and not xt are getting these cards whereas Radeon 7 wasn't really super wide availability so after recapping that we commented on how and he's got Junction temperature which if you don't know it's as reported in gpu-z it's the single hottest temperature out of all of the sensors on the die Radeon 7 had 64 of those that were checked and an edge temperatures the other number that AMD reports that is the literally the temperature at the edge of the die so it's going to be cooler on average and is what was traditionally worked with and presented before in that piece we talked about how Nvidia also has its GP reporting it's different from and AMD's so our original understanding was that Nvidia reported GPU edge temperature this is only that board partners told us several years ago but actually Nvidia emailed us to note that hey actually we report the average temperature so if you didn't know it because we didn't and so I assume most people don't the difference in what and video reports through gpu-z it's actually not an edge town as a correction that is an average of their sensors now unfortunately when we asked how many sensors do you have on the die and how was the averaging done things like that we couldn't get into any further details from Nvidia so all we know is that it's an average of all the sensors on side we don't know how many are on the die it does the count of sensors changes based upon the dye that's used but that's what we know so it's better than an edge temp and it's still not a junction temperature so you can't really linearly compare and eta2 Nvidia for thermals and this is I guess a really quick aside that deserves a different video but when trying to compare thermals between two different vendors all you really need is power this is a trap a lot of people fall into where they look at the temperature which is dependent upon things like the cooler and also obviously where the temperature is read from the only two that matter is how much power does it draw if it draws more power than its hotter and that's the end of the story the rest of it can be solved or fought between with different cooling solutions so anyway that was the update we wanted to give it's just it's a Nvidia reports an average of its sensors and then though further details were given Intel details count Lake so it's it's well known that Intel uses lakes to name its processors but unfortunately it has run out of lakes and now is referring to the processors by things like coffee lake whisky lake ice lake and comet-like many of these are names you probably know by now cascade lake can't forget that one everyone everyone remembers cascade lake from Computex comment lake is the 10th gen core series CPUs on 14 nanometer process it is not 10 it's got higher frequencies on the preceding whiskey Lake it's going to be used in notebooks and intel says that comet Lake will increase performance a this iterations used in notebooks Intel says count lake will increase performance by about 16 percent over the predecessor and then also noted that this is from the frequency increases but also a new memory controller and then memory improvements as well in terms of the maximum supported frequencies so even still with the near simultaneous launch of isolation count Lake and and advertently aims to confuse the buyer with two new product lines of different names at about the same time marketing is shaping up to target ice-like more towards gaming and notebooks underscored by the ice lake IGP inclusion well comet Lake will be more for consumer and busy line new processor names are also in confusing and include the Intel Core i7 one zero seven one zero u the core i7 one zero five one zero u the core i7 one zero two one zero you the 10110 you the ten five one zero why that was a close one the ten three one zero why and you can't forget the ten thousand two hundred and ten why or the 10110 why those are actually the names and of course following this generation for the next generation if you're wondering these names are kind of confusing they're getting really the numbers are getting longer and what's intel going to do next the next processors are going to be the Intel Core i7 8 6 7 5 3 or 9 and then the Douglas Adams Edition core i7 42 that's you heard it here first Nvidia adding integer scaling with new drivers so NVIDIA has buckled under community demand and has added integer scaling this is something that Intel to its credit did first just recently and Andy is under pressure now from the community to do the same so we'll probably see that from them eventually but what's what's going on here is basically if a game is run at a lower resolution of the display and GPU scaling which is an option is enabled in the Nvidia control panel the GPU will scale up the game to fit the screen although display scallion is actually the default meaning that the display handles the upscaling so a lot of users will probably never see this option unless they change that and Vidya currently uses regular old linear interpolation for resizing with Turing integer scaling aka nearest neighbor scaling being added as a new feature both of these names should be familiar to anyone that's resized a JPEG before in order to turn a few pixels into a lot of pixels linear interpolation averages pixels together to fill in the gaps resulting in a blurred image that's a compromise between speed and accuracy so for the comments section below you can keep your opinion on the best resizing algorithm to yourselves or you can validate yourselves by posting it nearest neighbor scaling just fills in the blanks with the color of the nearest pixel without any interpolation which is great for keeping pixel art sprites crisp and Nvidia's intended to use case for the integer scaling feature is running full screen pixel art games that don't natively support 4k like FTL Nvidia says adding the feature or the feature is enabled by quote a hardware accelerated programmable scaling filter available in terrain but nearest neighbor scaling is arguably the crudest possible resizing algorithm so hopefully the feature makes its way into older hardware in a recent second quarter 2019 earnings call and video CEO Jensen Huang known for greatest hits like it just works and it's beautiful look at it said that the he's all chips in basically on the ray tracing future of the company and also threw some shade at and the along the way so answering a question on how the new Nvidia super line is doing commercially quan said that it is a quote foregone conclusion that we're going to buy a new graphics card and it's going to last through two years three years four years to not have ray tracing is just crazy interestingly enough Juan did not mention that Nvidia cells GPS without ray tracing hardware so buying a TV without ray tracing is something you could do through Nvidia as well but you would be crazy to do it man said it himself technically Nvidia updated its driver stack to bring ray tracing two gtx cards previous generation 10 series cards i'll be it in a very limited fashion and with debatable merits and to be fair the non RT x hardware is more data center so different target dr. Lisa Sood so I cannot cross fire support this is something we asked AMD at their event for the rise in an a/v launch what's happening with crossfire and they said it'll have about the same support as previously so what's more recent is that it seems the collective industry is steadily moving away from the once envied Abel dual GPU setup and Vidya has long since abandoned its SLI marketing and ambitions although it still supports it and the company recently seems to downplay the idea with the launch of the the 700 series and their favorable performance in the mid-range segment the question regarding and these position on crossfire expectedly came up in short don't get your hopes up hot chips 2019 dr. Lisa soufeel dove the question regarding crossfire support from an attending press member and answered with quote to be honest the software is going faster than the hardware I would say that crossfire isn't a significant focus in other words we're giving up and that's really kind of a terrible answer on the reason why but also unexpected one AMD's our DNA architectural white paper is out now and it is very detailed we direct been checking it out well scroll scroll through some of it while I go through this coverage of it so with the and e7 nanometer Navi GPU is the arcs 5700 series and they also introduced its scalar our DNA graphics architecture from the start AMD has been pretty clear about its ambitions with our DNA and how the company plans for the architecture to proliferate from everything to smartphones and tablets all the way up to discrete GPU is with quote big navvy cards rumored to flesh out the upper end of the RX 5 series sometime later a recent white paper surfaced and this one details what's under the hood for the our DNA and future products and quote thanks to and these wide influence and extensive partnerships the our DNA architecture will roll out and eventually attach nearly every part of the industry the our DNA family will ultimately grow to include power constraints smartphone and tablet processors gaming consoles cloud gaming services and a full spectrum of gaming GPUs from low cost to the highest performance bringing the benefits of the rDNA architecture to millions of devices and people across the planet says the white paper so a lot of marketing there but don't let that get you down because it is a white paper so there's a lot of technical details too if you recall AMD and Samsung already announced that the two companies are planning to work together on phones coming up in a joint agreement to bring our DNA powered graphics to Samsung's SOC SoC is the power of the mobile devices in Samsung segment and that's slated to come to market around 2021 if not before and Andy is also working on custom solutions for the next-gen consoles from both Sony and Mike soft and they will further have a presence in cloud gaming via Google stadia and best we know Google stadia is currently running some variants of Vega 56 but presumably the service will migrate to an rDNA based solution and antes white paper our DNA's hypervisor agent is mentioned and how it relates to cloud gaming specifically there's a big block on that that says the hypervisor agent enables the GPU to be virtualized and shared between different operating systems most cloud gaming services live in data centers where virtualization is crucial from a security and operational standpoint while modern consoles focus on gaming many offer a rich suite of communication and media capabilities and benefit from virtualizing the hardware to deliver performance for all tasks and finally that paper outlines our DNA encompassing GPUs from low cost to the highest performance and he has been notably absent from the high end GPU market for some time and recent rumors like the Navi 21 and 23 rumors suggest that more might be coming so anyway check that white paper out if you want to learn more about navien how it works feel but also have some architecture discussions from when Navi was first really unveiled back at the press event for Rison and Navi I want to go back a couple months in time and check those videos finally Intel i5 9600 K price drops this is a short one but for those looking for a deal on Intel's case key chips we didn't particularly recommend any of the Intel i5 s after the AMD r5 series launched and started saturating the market that remains true with the 3000 series with a 9600 K it was an instance of it being technically better and a lot of the gaming tests but not we don't really feel great recommending it because unlike the eye 9 or even the 92 700 K now the i7 the 9600 K just doesn't offer enough and it's missing too many other important features so we do give the choice the r5 almost always that said Intel has dropped the price for at least the price has been dropped by the retailers it's now about $220 on Amazon at time of writing this news segment so that's 20 dollars lower than the previous price drop which had brought it down to $240 at launch it was about 200 for point of reference so until he's actually responding to pricing in some ways the 9600 K it's six cores it's six threads it lacks hyper-threading but it's overclockable and we talk about it most heavily in our 3600 review not the X but the 3600 if you wanna see the comparison between AMD the night 600k that's it for this one thank you for watching and subscribe for more you go to store documents nexus dotnet to sports directly like by buying one of these blue print shirts or our mod mats or page on Comm slash gamers Nexus I'll see you all next timeeveryone welcome back to another hardware news recap for the week we have some information on NVIDIA thermal monitoring Intel comet-like CPU details Nvidia adding integer scaling with the new drivers and then a couple of AMD stories like the rDNA white paper which is worth checking out and Lisa su commenting on crossfire support before that this video is brought to you by audible audible has a massive audio book library including content that talks computers and games audible has an entire series from the official Computer History Museum which we've actually toured in the past and can support as a leader in computer education audible also hosts the ultimate history of video games something I read back when researching GM content and can highly recommend for gaming and hardware enthusiasts audibles 30-day free trial can be unlocked at audible.com slash gamers Nexus or you can text gamers Nexus one word to 500 500 where you'll get a free audiobook and two audible originals or click the link in the description below so first one's really easy we talked about how AMD does junction temperature reading with the 5700 XT about a week ago and none of that was new that happened in February with the Radeon seven card launch but it was worth recapping because a lot of people with 500 XTC react and not xt are getting these cards whereas Radeon 7 wasn't really super wide availability so after recapping that we commented on how and he's got Junction temperature which if you don't know it's as reported in gpu-z it's the single hottest temperature out of all of the sensors on the die Radeon 7 had 64 of those that were checked and an edge temperatures the other number that AMD reports that is the literally the temperature at the edge of the die so it's going to be cooler on average and is what was traditionally worked with and presented before in that piece we talked about how Nvidia also has its GP reporting it's different from and AMD's so our original understanding was that Nvidia reported GPU edge temperature this is only that board partners told us several years ago but actually Nvidia emailed us to note that hey actually we report the average temperature so if you didn't know it because we didn't and so I assume most people don't the difference in what and video reports through gpu-z it's actually not an edge town as a correction that is an average of their sensors now unfortunately when we asked how many sensors do you have on the die and how was the averaging done things like that we couldn't get into any further details from Nvidia so all we know is that it's an average of all the sensors on side we don't know how many are on the die it does the count of sensors changes based upon the dye that's used but that's what we know so it's better than an edge temp and it's still not a junction temperature so you can't really linearly compare and eta2 Nvidia for thermals and this is I guess a really quick aside that deserves a different video but when trying to compare thermals between two different vendors all you really need is power this is a trap a lot of people fall into where they look at the temperature which is dependent upon things like the cooler and also obviously where the temperature is read from the only two that matter is how much power does it draw if it draws more power than its hotter and that's the end of the story the rest of it can be solved or fought between with different cooling solutions so anyway that was the update we wanted to give it's just it's a Nvidia reports an average of its sensors and then though further details were given Intel details count Lake so it's it's well known that Intel uses lakes to name its processors but unfortunately it has run out of lakes and now is referring to the processors by things like coffee lake whisky lake ice lake and comet-like many of these are names you probably know by now cascade lake can't forget that one everyone everyone remembers cascade lake from Computex comment lake is the 10th gen core series CPUs on 14 nanometer process it is not 10 it's got higher frequencies on the preceding whiskey Lake it's going to be used in notebooks and intel says that comet Lake will increase performance a this iterations used in notebooks Intel says count lake will increase performance by about 16 percent over the predecessor and then also noted that this is from the frequency increases but also a new memory controller and then memory improvements as well in terms of the maximum supported frequencies so even still with the near simultaneous launch of isolation count Lake and and advertently aims to confuse the buyer with two new product lines of different names at about the same time marketing is shaping up to target ice-like more towards gaming and notebooks underscored by the ice lake IGP inclusion well comet Lake will be more for consumer and busy line new processor names are also in confusing and include the Intel Core i7 one zero seven one zero u the core i7 one zero five one zero u the core i7 one zero two one zero you the 10110 you the ten five one zero why that was a close one the ten three one zero why and you can't forget the ten thousand two hundred and ten why or the 10110 why those are actually the names and of course following this generation for the next generation if you're wondering these names are kind of confusing they're getting really the numbers are getting longer and what's intel going to do next the next processors are going to be the Intel Core i7 8 6 7 5 3 or 9 and then the Douglas Adams Edition core i7 42 that's you heard it here first Nvidia adding integer scaling with new drivers so NVIDIA has buckled under community demand and has added integer scaling this is something that Intel to its credit did first just recently and Andy is under pressure now from the community to do the same so we'll probably see that from them eventually but what's what's going on here is basically if a game is run at a lower resolution of the display and GPU scaling which is an option is enabled in the Nvidia control panel the GPU will scale up the game to fit the screen although display scallion is actually the default meaning that the display handles the upscaling so a lot of users will probably never see this option unless they change that and Vidya currently uses regular old linear interpolation for resizing with Turing integer scaling aka nearest neighbor scaling being added as a new feature both of these names should be familiar to anyone that's resized a JPEG before in order to turn a few pixels into a lot of pixels linear interpolation averages pixels together to fill in the gaps resulting in a blurred image that's a compromise between speed and accuracy so for the comments section below you can keep your opinion on the best resizing algorithm to yourselves or you can validate yourselves by posting it nearest neighbor scaling just fills in the blanks with the color of the nearest pixel without any interpolation which is great for keeping pixel art sprites crisp and Nvidia's intended to use case for the integer scaling feature is running full screen pixel art games that don't natively support 4k like FTL Nvidia says adding the feature or the feature is enabled by quote a hardware accelerated programmable scaling filter available in terrain but nearest neighbor scaling is arguably the crudest possible resizing algorithm so hopefully the feature makes its way into older hardware in a recent second quarter 2019 earnings call and video CEO Jensen Huang known for greatest hits like it just works and it's beautiful look at it said that the he's all chips in basically on the ray tracing future of the company and also threw some shade at and the along the way so answering a question on how the new Nvidia super line is doing commercially quan said that it is a quote foregone conclusion that we're going to buy a new graphics card and it's going to last through two years three years four years to not have ray tracing is just crazy interestingly enough Juan did not mention that Nvidia cells GPS without ray tracing hardware so buying a TV without ray tracing is something you could do through Nvidia as well but you would be crazy to do it man said it himself technically Nvidia updated its driver stack to bring ray tracing two gtx cards previous generation 10 series cards i'll be it in a very limited fashion and with debatable merits and to be fair the non RT x hardware is more data center so different target dr. Lisa Sood so I cannot cross fire support this is something we asked AMD at their event for the rise in an a/v launch what's happening with crossfire and they said it'll have about the same support as previously so what's more recent is that it seems the collective industry is steadily moving away from the once envied Abel dual GPU setup and Vidya has long since abandoned its SLI marketing and ambitions although it still supports it and the company recently seems to downplay the idea with the launch of the the 700 series and their favorable performance in the mid-range segment the question regarding and these position on crossfire expectedly came up in short don't get your hopes up hot chips 2019 dr. Lisa soufeel dove the question regarding crossfire support from an attending press member and answered with quote to be honest the software is going faster than the hardware I would say that crossfire isn't a significant focus in other words we're giving up and that's really kind of a terrible answer on the reason why but also unexpected one AMD's our DNA architectural white paper is out now and it is very detailed we direct been checking it out well scroll scroll through some of it while I go through this coverage of it so with the and e7 nanometer Navi GPU is the arcs 5700 series and they also introduced its scalar our DNA graphics architecture from the start AMD has been pretty clear about its ambitions with our DNA and how the company plans for the architecture to proliferate from everything to smartphones and tablets all the way up to discrete GPU is with quote big navvy cards rumored to flesh out the upper end of the RX 5 series sometime later a recent white paper surfaced and this one details what's under the hood for the our DNA and future products and quote thanks to and these wide influence and extensive partnerships the our DNA architecture will roll out and eventually attach nearly every part of the industry the our DNA family will ultimately grow to include power constraints smartphone and tablet processors gaming consoles cloud gaming services and a full spectrum of gaming GPUs from low cost to the highest performance bringing the benefits of the rDNA architecture to millions of devices and people across the planet says the white paper so a lot of marketing there but don't let that get you down because it is a white paper so there's a lot of technical details too if you recall AMD and Samsung already announced that the two companies are planning to work together on phones coming up in a joint agreement to bring our DNA powered graphics to Samsung's SOC SoC is the power of the mobile devices in Samsung segment and that's slated to come to market around 2021 if not before and Andy is also working on custom solutions for the next-gen consoles from both Sony and Mike soft and they will further have a presence in cloud gaming via Google stadia and best we know Google stadia is currently running some variants of Vega 56 but presumably the service will migrate to an rDNA based solution and antes white paper our DNA's hypervisor agent is mentioned and how it relates to cloud gaming specifically there's a big block on that that says the hypervisor agent enables the GPU to be virtualized and shared between different operating systems most cloud gaming services live in data centers where virtualization is crucial from a security and operational standpoint while modern consoles focus on gaming many offer a rich suite of communication and media capabilities and benefit from virtualizing the hardware to deliver performance for all tasks and finally that paper outlines our DNA encompassing GPUs from low cost to the highest performance and he has been notably absent from the high end GPU market for some time and recent rumors like the Navi 21 and 23 rumors suggest that more might be coming so anyway check that white paper out if you want to learn more about navien how it works feel but also have some architecture discussions from when Navi was first really unveiled back at the press event for Rison and Navi I want to go back a couple months in time and check those videos finally Intel i5 9600 K price drops this is a short one but for those looking for a deal on Intel's case key chips we didn't particularly recommend any of the Intel i5 s after the AMD r5 series launched and started saturating the market that remains true with the 3000 series with a 9600 K it was an instance of it being technically better and a lot of the gaming tests but not we don't really feel great recommending it because unlike the eye 9 or even the 92 700 K now the i7 the 9600 K just doesn't offer enough and it's missing too many other important features so we do give the choice the r5 almost always that said Intel has dropped the price for at least the price has been dropped by the retailers it's now about $220 on Amazon at time of writing this news segment so that's 20 dollars lower than the previous price drop which had brought it down to $240 at launch it was about 200 for point of reference so until he's actually responding to pricing in some ways the 9600 K it's six cores it's six threads it lacks hyper-threading but it's overclockable and we talk about it most heavily in our 3600 review not the X but the 3600 if you wanna see the comparison between AMD the night 600k that's it for this one thank you for watching and subscribe for more you go to store documents nexus dotnet to sports directly like by buying one of these blue print shirts or our mod mats or page on Comm slash gamers Nexus I'll see you all next time\n"