AMD Officially Launches Bristol Ridge Processors And Zen-Ready AM4 Platform

24 Comments

1. 

   Tim Verry on September 8, 2016 at 4:44 am

   Hmm looks like Eurogamer

   Hmm looks like Eurogamer benchmarked the i5-6500 in Cinebench R11.5 single thread at 1.68. Their benchmarking rig is not identical though:

   "So let's talk about our test system. We bought the Core i5 6500 ourselves from Amazon, running the system from an MSI Z170A Gaming M5 motherboard, with all tests run from a 512GB Crucial MX100 SSD. Our all-important DDR4 came in the form of two 4GB sticks of Corsair Vengeance LPX DDR4, rated for 3000MHz. Curiously, using the XMP profile for these sticks applied a mild two per cent CPU overclock, so we overrode that and increased the RAM multiplier instead, giving us 3066MHz. Overvolting the RAM to 1.35v allowed us to go further, easily hitting a stable 3200MHz. These new modules would allow us to test a whole spectrum of DDR4 frequencies and to get an idea of how faster RAM impacts gaming performance on this i5 quad."

   http://www.eurogamer.net/articles/digitalfoundry-2016-intel-skylake-core-i5-6500-review

   Reply
2. 

   Anonymous on September 8, 2016 at 5:13 am

   With the OEMs currently the
   With the OEMs currently the only ones with AM4 and Bristol Ridge SKUs! One can imagine just how much Bristol Ridge will be gimped of performance potential in that hands of the very same OEMs that produced the wonderfully gimped Carrizo Laptop abominations of the first Excavator based designs, with all that single channel and low wattage limitations that really made Carrizo take to the sky and soar the way a brick does! I have very little trust any any OEM and AMD’s latest product considering the past performance as a good indicator of what Bristol Ridge will receive in the hands of those OEM’s! Color me Jade but I’m waiting for the home system builder market to maybe come to the rescue of Bristol Ridge in the interval before Zen becomes fully available without a long wait for the available Zen SKU supplies to meet demand.

   OEM’s are why the PC market is in the crapper in the first place, that and the laptop market as of the past few years producing such overpriced/underpowered duds, and Windows 10 is not a big help also following on the heels of windows 8.

   Reply
   • 

     Anonymous on September 8, 2016 at 6:18 am

     AMD is not without fault
     AMD is not without fault themselves. Only recently they started to fine tune their designs for power efficiency (for example cutting PCIe lanes to 8 which is still plenty for the segment their CPUs are in).

     Reply
     • 

       Anonymous on September 8, 2016 at 5:24 pm

       No, just No! The post you
       No, just No! The post you replied to does not deal with PCIe lanes, it Deals with OEM’s and the lack of Trust when OEM’s are using AMD’s SKUs in their OEM PC/Laptop products. Thankfully there will be Bristol Ridge and AM4 SKUs for the home system builder market, at a later time! I hope that the home system builder market gets some BR/AM4 availability before the Zen SKUs are available, we all know that Zen will be limited in availability when first released! And the OEM’s can Kiss my A$$, I’m building my own PC this time around. I’ll just have to wait for a Zen/Polaris interposer based laptop SKU with at least one HBM2 stack of 4GB of HBM2 that no laptop OEM can gimp of channel capacity to hurt the APU’s integrated Polaris Graphics performance!

       Reply
3. 

   Anonymous on September 8, 2016 at 6:24 am

   “GCN 3.0-based Radeon R7
   “GCN 3.0-based Radeon R7 graphics (that support VP9 and HEVC acceleration)”
   “HEVC acceleration is subject to inclusion/installation of compatible HEVC players.”

   What are the compatible players?
   Intel at least said what power consumption we can expect playing VP9 Yutube videos using Chrome browser.

   Reply
4. 

   JohnGR on September 8, 2016 at 9:10 am

   These chipsets look like OEM
   These chipsets look like OEM chipsets. I am expecting BIOS option for limiting the graphics card at 4X and PCIe SSDs to use maximum 2 PCIe lanes if you want to have both of them AND a graphics card. And yes there is an Athlon in there that needs an extra graphics card.

   Reply
5. 

   Anonymous on September 8, 2016 at 1:06 pm

   Look at the footnotes. What’s
   Look at the footnotes. What’s going on here? Quote “…the 7th Generation AMD A12-9800 scored 3483.25, while the Intel Core i5-6500 scored 3702 for a benchmark score comparison of 3483.25/3466.5 = 1.00X or 100%…”. If “Intel Core i5-6500 scored 3702” then that is 106% faster than A12-9800. The question is where does that “3466.5” came from?

   Reply
   • 

     Tim Verry on September 8, 2016 at 3:07 pm

     Yeah I wasn’t sure what they

     Yeah I wasn't sure what they were going for there heh.

     Reply
   • 

     devin w lane on September 9, 2016 at 8:06 pm

     Morons…. 3702-3483=219..
     Morons…. 3702-3483=219.. 219/3483=.062 or 6.2% increase over the and, in that category, but amd posted a 99.4% increase compared with the i5 in the other benchmark. Not sure where they got the 3466 from, probably a key error, and not the only one on the page. But that flips. The comparisons. Should read that the amd performed 98.9% as well in pcmark and 99.4% better in 3dmark

     Reply
6. 

   Anonymous on September 8, 2016 at 1:16 pm

   AMD’s “quad core” in APUs are
   AMD’s “quad core” in APUs are not true quad cores, thus the performance is not even comparable to Intel’s true quad cores from CPU performance point of view. Those slides are also misleading. AMD is comparing their new A12-9800 with Intel Core i5-6500 using PCMark8 “Accelerated” which uses OpenCL acceleration. This means its not a pure CPU oriented system test or benchmark. AMD’s integrated Radeon GPU is generally quicker than Intel’s HD Graphics (excluding Intel’s Iris Pro) when it comes to Direct3D, OpenGL and OpenCL performance.

   Reply
   • 

     anon on September 8, 2016 at 11:13 pm

     cpus today are not pure cpus.
     cpus today are not pure cpus. they come with graphics. so why not take advantage of the better integrated gpu by running an accelarated benchmark? would it be wrong to compare the two processors in video encoding where quicksync works?

     Reply
     • 

       Anonymous on September 9, 2016 at 5:16 pm

       AMD has been touting
       AMD has been touting integrated graphics acceleration for general purpose usage since the APU’s inception, first thru OpenCL (mainly used in HPC nowadays) and then later AMD created HSA (which is still not in wide use at all because of hardware compliance requirements). The main problem is that a majority of everyday applications (office, legacy, productivity, etc) and even gaming software used by most PC consumers (average Joes) do not use OpenCL. Thus that is why this advertisement is misleading.

       Reply
       • 

         Anonymous on September 10, 2016 at 7:06 pm

         OpenOffice is used by
         OpenOffice is used by (average Joes), even brain Dead Joe Six-Packs, and OpenOffice accelerates spreadsheets via OpenCL on the GPU. And Average Joe is not the one doing the programming for any of the applications that any average Joe/Jane uses in the first place. So WTF are you talking about. And HSA(heterogeneous system architecture) was not invented by AMD, it’s been around in computing for decades with the mainframe computers using vector processors long before vector processors where taken and used to make GPUs, with GPU’s having some extra specialized hardware added to accelerate geometry and raster operations. HSA is of great use to the programming profession who writes the applications that use the HSA aware OSs/API/Hardware that make use of the HSA compute on GPUs/DSPs/other processors that (average Joes) use, whether average Joe or Jane know it or not.

         OS’s that utilize GPU/Other processor hardware that use OpenCL/other APIs(Vulkan, HSAIL) can all be said to have some form of HSA abilities, as all the acronym HSA really means is the ability of an OS’s/API’s and computing system hardware/drivers that designed to be able to are able to use GPUs, FPGAs, DSPs, etc. for all forms of compute. If you look at a GPU, well that GPU’s hardware does not Know that it’s a GPU, the GPU’s hardware can add, subtract, multiply, and divide, and do shift/other, operations on groups of binary ones and zeros when its op codes tell it to the work on the GPU’s processors(Shaders).

         The GPU does not know that it has “shaders” it just knows how to execute the op codes that run on the shader(Processors), the software tells the shaders to what to do and the shaders(Processors) do the adding/subtracting/whatever! So what if the data that is executed on the GPU’s shaders was actually Ones and Zeros to be added and subtracted for a spread sheet and not an image, the GPU does not know the difference and its GPU shader processors just follow what the op codes tell it to do with the binary ones and zeros. That’s what HSA is it’s just doing any form of processing on the GPU, or DSP, Or other ASIC above and beyond that ASIC’s original intended usage. CPU’s can do rendering they are not very fast at rendering but still they can do rendering! CPU’s do not know that they are CPUs and not GPUs so what’s the big mystery about HSA.

         Don’t let AMD’s marketing fool you into thinking that AMD invented HSA, as heterogeneous systems have been in use in computing for a very long time. AMD may have embraced the design tenets around what is called HSA today, but the membership of the HSA foundation should be an indicator that the Computing Sciences Ideas behind HSA have been around for decades and many more companies are using HSA than just AMD. Just because an APU/SOC does not comply with the HSA foundation’s “HSA 1.0” specification does not mean that any companies’ APUs/SOCs does not have any HSA ability at all, as even Intel’s SOCs since IvyBridge have had OpenCL enabled drivers, and there will be Vulkan ability there also to some degree on Intel’s integrated GPUs, as both Vulkan and OpenCL utilize SPIR-V, and SPIR-V has a lot of the very same/similar reach of the HSA foundations’ HSAIL specification. Hell even ARM Holdings, a founding member of the HSA foundation, is using SPIR-V more than HSAIL for ARM holdings’ GPGPU compute at the moment, and that new ARM Holdings Mali/Bifrost GPU micro-architecture is very async compute enabled compared to AMD’s GCN micro-architecture.

         Reply
         • 

           Anonymous on September 11, 2016 at 9:00 am

           Majority of ordinary folks
           Majority of ordinary folks use Microsoft stuff like Microsoft Windows and Microsoft Office. The only place where OpenOffice is used mostly is on systems using Linux and BSD operating systems where Microsoft Office is not available. Meanwhile most everyday applications besides Microsoft Office mainly does not use OpenCL or simply has OpenCL as an option only (disabled by default). That’s because OpenCL support depends on the hardware, thus not every integrated GPU can support all OpenCL features. Example http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/71760-intel-skylake-i5-6500-i5-6400-i3-6100-review-8.html major flip flop there, Blender could not use Intel’s integrated GPU and likewise Handbrake could not use AMD’s integrated GPU. Additionally do not confuse “heterogenous” systems with HSA. Actually HSA was AMD’s creation, while OpenCL was Apple’s Creation. Why? Because AMD was the original HSA founding principle. Also because currently only AMD has the first compliant HSA hardware in the market and that was Carrizo (and probably Bristol Ridge also which was based on Carrizo). Meanwhile the rest of the HSA foundation members do not have it (you can try searching and you will find none, other hardware are mostly “heterogenous” thru OpenCL). This is why HSA is very rarely seen.

           Reply
           • 

             Anonymous on September 13, 2016 at 2:21 am

             HSA is a marketing term for
             HSA is a marketing term for AMD’s version of a heterogeneous system(HS), look at the old supercomputers that used vector processors they where HS, they had CPUs and Vector processors. The Vector processors(2) of old where made into the GPUs we see today. Any Blender SD work that requires high polygon count meshes chokes Intel’s Integrated GPUs, I had to get an HP probook with an AMD GPU just to be able to work with high polygon count mesh modeling in Blender’s 3d editing mode that uses OpenGL in the editing mode to render the meshes so I can edit them. Intel’s graphics are tuned for low polygon count gaming scenes and can not handle any high polygon count 3d meshes without bogging down in the edit mode on Blender 3d. Only AMD’s and Nvidia’s GPUs have enough Shader cores to be able to handle High polygon count meshes in Blender’s 3D editing mode, while Intel’s low numbers of shader processors are only good for low polygon count meshes used in finished games. Intel’s GPUs are not good for game creation where things are done first on high polygon count high detail meshes to produce the high resolution textures and then skinned onto low polygon count gaming meshes to run on the games.

             I’ll bet that you thing that Intel’s HyperThreading is a real thing when HyperThreading is just a fancy marketing Brand Name for Intel’s version of SMT(Simultaneous multi-threading), and Intel did not Invent SMT(1). The same goes HSA(heterogeneous system architecture), AMD did not invent the first heterogeneous systems, that was done decades ago. AMD’s Zen processor will have SMT abilities, but you really need to fine tune you marketing BS RADAR and learn the proper generic computing sciences for these technologies that have been around before there was an Intel or an AMD.

             (1)

             “Historical implementations

             While multithreading CPUs have been around since the 1950s, simultaneous multithreading was first researched by IBM in 1968 as part of the ACS-360 project.[1] The first major commercial microprocessor developed with SMT was the Alpha 21464 (EV8). This microprocessor was developed by DEC in coordination with Dean Tullsen of the University of California, San Diego, and Susan Eggers and Henry Levy of the University of Washington. The microprocessor was never released, since the Alpha line of microprocessors was discontinued shortly before HP acquired Compaq which had in turn acquired DEC. Dean Tullsen’s work was also used to develop the Hyper-threading (Hyper-threading technology or HTT) versions of the Intel Pentium 4 microprocessors, such as the “Northwood” and “Prescott”.”(1)

             (1)

             https://en.wikipedia.org/wiki/Simultaneous_multithreading

             (2)

             https://en.wikipedia.org/wiki/Vector_processor

             P.S. Blender runs on Windows and Linux, and so does OpenOffice, Gimp, Krita, etc.! Pleease research before commenting! I’ll be on Linux OS laptop come 2020 hopefully on a Zen/Polaris APU with plenty of Shaders to run those high polygon mesh models!

             Reply
             • 

               Anonymous on September 13, 2016 at 12:10 pm

               You are wrong again and
               You are wrong again and again. As mentioned earlier HSA requires compliant hardware and was created by AMD. Read http://www.anandtech.com/show/10387/hsa-11-specification-launched-multi-vendor-support quiote “Originally envisioned by AMD, HSA has been the cornerstone of their efforts to develop a fully functional ecosystem for heterogeneous hardware and software. Define a standard, make it easy(ier) for developers to create software around it, and, if all goes well, AMD’s big bet on GPU technology made almost 10 years ago will pay off.

               However while AMD was the birthplace of what would become HSA, the standard as a whole has been about more than just one company. Which is why HSA has been under development of the HSA Foundation since 2012. ” and quote “ARM for their part is following an OpenCL-centric software strategy for exposing the heterogeneous aspects of their hardware to developers. The interesting thing about the ARM implementation is that they have gone above and beyond the basic aspects of OpenCL 2.0, offering finer-grained sync that OpenCL requires at a minimum. Finer-grained sync that would otherwise be something better suited for HSA. As a result part of the HSA Foundation’s efforts are focused on showcasing the additional benefits of HSA over OpenCL 2.0, to entice hardware manufacturers and developers into using HSA.”. Also http://www.anandtech.com/show/10375/arm-unveils-bifrost-and-mali-g71/4 quote “With yesterday’s announcement of the HSA 1.1 specification, I went back to ARM to ask them whether the new specification impacts the company’s heterogenous compute plans at all, especially given that their architecture doesn’t support the 1.0 standard. As it turns out, ARM is going a route very similar to AMD’s ROCm platform: while the company isn’t utilizing the HSAIL – and thus in the strictest sense isn’t a complete HSA platform – they are using the HSA standard in the development of their hardware.

               At a hardware level, the HSA specification standardizes a number of aspects of the hardware for common interoperability and easier programming purposes, including signals, queues, floating point number handling, and other, low-level minutiae about how heterogeneous execution should work. This is separate from the HSAIL, which is more concerned with the software aspects of heterogeneous programming, and though helpful, is not necessary for heterogeneous compute. As a result while Mali-G71 is technically not an HSA platform, in practice it is HSA hardware, using the HSA specification as a means to offer a common and well understood execution model for heterogeneous compute. So ARM is very much on-board with HSA – and is essentially supplying one of the first non-AMD HSA 1.1 hardware designs – even if they’re not using HSAIL itself.”

7. 

   Anonymous Nvidia User on September 8, 2016 at 6:11 pm

   AMD is predictable. Don’t run
   AMD is predictable. Don’t run the Convenience for the Akhalie because it shows their CPU to be very lacking in single core performance.

   http://www.cpu-monkey.com/en/compare_cpu-intel_core_i5_6500-524-vs-intel_core_i5_4690-356

   From this bench Skylake scores 1.83 vs AMD’s 1.21. So the Skylake is up to a 51% better processor than the AMD. Silence from the AMD crowd.

   As far as the pcmark8 goes the only benches listed on 3dmark site for a10 is the a10-7850k which averages around 4000 pts.

   http://www.3dmark.com/search#/?mode=basic&url=/proxycon/ajax/search/cpuname/pcm8hme/A/AMD%20A10-7850K&cpuName=AMD A10-7850K

   According to this a10-8850 is 100mhz faster and has igpu added.
   Performance expected to be very similar.

   http://www.cpu-world.com/Compare/486/AMD_A10-Series_A10-7850K_vs_AMD_A10-Series_A10-8850K.html

   So extrapolating scores 3483.25 vs 4000 for about 15% less performance than the a10-8850.

   There I filled in the blanks that AMD didn’t want filled in. Now you know why. If AMD is presenting their benchmarks in the best light only you should be very worried. Zen isn’t all it’s claimed to be.

   Reply
   • 

     Anonymous Nvidia User on September 8, 2016 at 6:24 pm

     Stupid Google autocorrect.
     Stupid Google autocorrect. Convenience for the Akhalie is supposed to be Cinebench for the Skylake.

     Sorry meant to put if they start releasing Zen benchmarks with blanks not filled in just like with “Bristol Ridge”, you should run away from it. Benchmarks are meant to expose strengths and weaknesses. By excluding results you find unacceptable, it is misleading the potential buyer.

     Reply
     • 

       Tim on September 8, 2016 at 7:19 pm

       ah heh I was like akhile
       ah heh I was like akhile comvenience? 0.o

       Yeah I saw the 1.83 result but I was not sure how reloable thay site is/was. I posted above that Eurogamrr benchmarked it at 1.68. Hopefully Zen will be more competitive or even beat that number. I just wish Bristol Ridge had been what Bulldozer was not (ie BR was what we got as the first Bulldozer processors). They seem to have finally gotten the clockspeed issues sorted out just in time to retire this cmt line of designs. Better late than never i suppose heh.

       Reply
   • 

     Anonymous on September 9, 2016 at 3:45 am

     Bristol Ridge is not zen, it
     Bristol Ridge is not zen, it is still excavator cores. The upcoming summit ridge is zen.

     Reply
   • 

     Anonymous on September 10, 2016 at 7:44 pm

     You have got to be one of
     You have got to be one of Chizow’s sockpuppets! And BR/excavator is NOT Zen, with Zen having way more Full Fat core resources and SMT! And Zen has no core module decoder/FPU sharing. That Zen 4 core complex is 4 Full Fat x86 SMT2 cores! Read and Learn:

     “AMD Zen Microarchiture Part 2: Extracting Instruction-Level Parallelism”

     http://www.anandtech.com/show/10591/amd-zen-microarchiture-part-2-extracting-instructionlevel-parallelism

     Reply
   • 

     Kraaketaer on September 11, 2016 at 11:26 am

     From my experience, there are
     From my experience, there are few AMD fans out there (myself included) that don’t readily accept that AMDs current architectures (the various lawnmowers and such) lag significantly behind Intel in IPC and single threaded performance. AMD itself of course needs to play to its strengths, and as such touts scenarios where they do well (anything GPU-assisted, as well as a few other measures).

     OTOH, GPU acceleration is growing in relevance – for anything from browsers to video and photo editing to spreadsheets. Having a strong iGPU is a good thing, no doubt. If it wasn’t, Intel wouldn’t have made the strides they have in recent years. But at the same time, having a strong CPU to begin with is hard to beat.

     Which is why AMD is working on Zen. They wouldn’t be making an entirely new architecture if they didn’t acknowledge that their current one is a bad fit for today’s software. The 2 cores/1 module approach was a bet on software evolving alongside the architecture in a way it simply didn’t turn out to do. That’s not AMDs fault, but they need to adapt – and they are doing so in a big way.

     Reply
   • 

     Anonymous on September 11, 2016 at 4:52 pm

     These Bristol Ridge chips
     These Bristol Ridge chips aren’t Zen.

     But that doesn’t matter. You talked shit about AMD. Mission accomplished, congratulations. Do you feel better?

     Reply
8. 

   Robert Beres on September 8, 2016 at 7:59 pm

   I don’t care if Intel is
   I don’t care if Intel is better or AMD. I will tell you this much my money will always go to the one that will save me money in the Electricity department. and still capable of running games without terrible ping issues or freezing while playing. so if AMD puts out a CPU capable of saving me the bottom line in money in all aspects well I guess Intel will learn that there are more people on a budget these days then there were 10 years ago. taxes are being raised, and forced to pay for crap you don’t need. yep the democratic way has begun.

   Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.