AMD Zen Diagram Leaked and Analysis

Subject: Processors | April 27, 2015 - 06:06 PM |
Tagged: Zen, Steamroller, Kaveria, k12, Excavator, carrizo, bulldozer, amd

There are some pretty breathless analysis of a single leaked block diagram that is supposedly from AMD.  This is one of the first indications of what the Zen architecture looks like from a CPU core standpoint.  The block diagram is very simple, but looks in the same style as what we have seen from AMD.  There are some labels, but this is almost a 50,000 foot view of the architecture rather than a slightly clearer 10,000 foot view.

There are a few things we know for sure about Zen.  It is a clean sheet design that moves away from what AMD was pursuing with their Bulldozer family of cores.  Zen gives up CMT for SMT support for handling more threads.  The design has a cluster of four cores sharing 8 MB of L3 cache, with each core having access to 512 KB of L2 cache.  There is a lot of optimism that AMD can kick the trend of falling more and more behind Intel every year with this particular design.  Jim Keller is viewed very positively due to his work at AMD in the K7 through K8 days, as well as what he accomplished at Apple with their ARM based offerings.

zen.jpg

One of the first sites to pick up this diagram wrote quite a bit about what they saw.  There was a lot of talk about, “right off the bat just by looking at the block diagram we can tell that Zen will have substantially higher single threaded performance compared to Excavator and the Bulldozer family.”  There was the assumption that because it had two 256-bit FMACs that it could fuse them to create a single 512 bit AVX product.

These assumptions are pretty silly.  This is a very simple block diagram that answers few very important questions about the architecture.  Yes, it shows 6 int pipelines, but we don’t know how many are address generation vs. execution units.  We don’t know how wide decode is.  We don’t know latency to L2 cache, much less how L3 is connected and shared out.  So just because we see more integer pipelines per core does not automatically mean, “Da, more is better, strong like tractor!”  We don’t know what improvements or simplifications we will see in the schedulers.  There is no mention of the front-end other than Fetch and Decode.  How about Branch Prediction?  What is the latency for the memory controller when addressing external memory?

Essentially, this looks like a simplified way of expressing to analysts that AMD is attempting to retain their per core integer performance while boosting floating point/AVX at a similar level.  Other than that, there is very little that can be gleaned from this simple block diagram.

Other leaks that are interesting concerning Zen are the formats that we will see these products integrated into.  One leak detailed a HPC aimed APU that features 16 Zen cores with 32 MB of L3 cache attached to a very large GPU.  Another leak detailed a server level chip that will support 32 cores and will be seen in 2P systems.  Zen certainly appears to be very flexible, and in ways it reminds me of a much beefier Jaguar type CPU.  My gut feeling is that AMD will get closer to Intel than it has been in years, and perhaps they can catch Intel by surprise with a few extra features.  The reality of the situation is that AMD is far behind and only now are we seeing pure-play foundries start to get even close to Intel in terms of process technology.  AMD is very much at a disadvantage here.

Still, the company needs to release new, competitive products that will refill the company coffers.  The previous quarter’s loss has dug into cash reserves, but AMD is still stable in terms of cash on hand and long term debt.  2015 will see new GPUs, an APU refresh, and the release of the new Carrizo parts.  2016 looks to be the make or break year with Zen and K12.

Edit 2015-04-28:  Thanks to SH STON we have a new slide that has been leaked from the same deck as this one.  This has some interesting info in that AMD may be going away from exclusive cache designs.  Exclusive was a good idea when cache was small and expensive, as data was not replicated through each level of cache (L1 was not replicated in L2 and L2 was not replicated in L3).  Intel has been using inclusive cache since forever, where data is replicated and simpler to handle.  Now it looks like AMD is moving towards inclusive.  This is not necessarily a bad thing as the 512 KB of L2 can easily handle what looks to be 128 KB of L1 and the shared 8 MB of L3 cache can easily handle the 2 MB of L2 data.  Here is the link to that slide.

zen2.jpg

The new slide in question.

Source: AMD

Godavari rumours abound

Subject: General Tech | April 27, 2015 - 01:04 PM |
Tagged: Summit Ridge, rumours, Godavari, Bristol Ridge, amd

This morning DigiTimes is reporting on a unconfirmed rumour that AMD's new APU, Godavari, will be arriving towards the end of May of this year.  This goes along with the leak that WCCFtech reported on this weekend, they have information that the chip will be a Kaveri design with up to four Steamroller cores, a GCN 1.1 base GPU with up to 512 stream processors and a dual-channel DDR3 memory in an FM2+ socket.  If their information is accurate you can expect to see models with 65W or 95W TDPs and boost clocks in the 4GHz range depending on the model.  Also worth noting is the rumour that AMD has placed orders with ASMedia Technology for USB 3.1 controller ICs for release in September.

index.png

"AMD will launch Godavari series APUs at the end of May to compete with Intel's Broadwell and Skylake platforms, according to Taiwan-based supply chain makers."

Here is some more Tech News from around the web:

Tech Talk

Source: DigiTimes

AMD and QNAP get NASty

Subject: Storage | April 23, 2015 - 03:39 PM |
Tagged: TVS-463 8G, qnap, NAS, amd

The QNAP TVS-463 8G is powered by an AMD GX-424CC, part of the Steppe Eagle family of SoCs which includes a Mullin's based Radeon R5E GPU.  There are several models ranging from the entry level which sports only 4GB of RAM, which can be expanded to 16GB with the review model TechPowerUp recieved sitting in the middle at 8GB.  You can install up to four 2.5" or 3.5" SATA3 disks in a variety of RAID configurations, the NAS ships empty so you will need to provide your own drives.  It is a little expensive, just over $800, which includes the internal PSU and the built in OS to allow you to activate your NAS via the web with a simple command.  It has two Gigabit ports with LACP support and you can even pick up an expansion card to increase it to 10GbE, read the full review to get an idea just how capable this NAS is.

nas_quarter.jpg

"QNAP has for the first time used an AMD CPU with one of their NAS offerings. The new series is codenamed TVS-x63, and today, we will evaluate the TVS-463, which, as its model number implies, can take up to four HDDs. It is also 10GbE ready through an optional expansion card."

Here are some more Storage reviews from around the web:

Storage

Source: techPowerUp

The Linux AMDGPU for R9 285 arrives

Subject: General Tech | April 21, 2015 - 03:07 PM |
Tagged: tonga, linux, carrizo, AMDGPU, amd

It will not be officially rolled in until kernel 4.2 but you can currently grab the new binary blob by following the links from Phoronix.  This new AMDGPU kernel driver will be used by both the full open-source driver and the Catalyst driver provided officially by AMD and provide support not only for the R9 285 but upcoming families as well.  There is still some development to be done as AMD's Alex Deucher told Phoronix that this initial code lacks power management features for Tonga but that will be addressed shortly.

index.jpg

"At long last the source code to the new AMDGPU driver has been released! This is the new driver needed to support the Radeon R9 285 graphics card along with future GPUs/APUs like Carrizo. Compared to the existing Radeon DRM driver, the new AMDGPU code is needed for AMD's new unified Linux driver strategy whereby the new Catalyst driver will be isolated to being a user-space binary blob with both the full open-source driver and the Catalyst driver using this common AMDGPU kernel driver."

Here is some more Tech News from around the web:

Tech Talk

Source: Phoronix

AMD's Lisa Su Expects Windows 10 to Launch in July

Subject: General Tech | April 21, 2015 - 07:00 AM |
Tagged: windows 10, windows, microsoft, amd

The CEO of AMD is an unexpected, but probably very accurate, source when it comes to knowing the Windows 10 release date. First off, the news broke on a quarterly earnings call. When you make a statement on those, you have a strong legal obligation to be telling the truth according to the knowledge that you have at the time. Also, as a major hardware vendor of CPUs and GPUs, her company would have been notified by Microsoft so that they could plan development of graphics drivers and so forth. It also aligns with the “Summer” announcement made last month by Microsoft.

She believes that Windows 10 is set to launch in July.

Windows95logo.png

Of course, this led to a flurry of comments that claim three months will not be enough time to bake a successful product. Others, naturally, claim that Microsoft has been developing software for long enough to know that they can finish their product in three months. Still others shrug and say, “Yeah, you both make sense. I'm going to go play some Grand Theft Auto.”

One aspect that I don't see mentioned enough is that Microsoft has multiple projects and teams on the go, and we only see a fraction of what is being done in our Insider branch. Despite the narrative that Microsoft wishes to avoid another Windows 8 fiasco and they want their users to guide development, they have alluded that a major reason for the Insider program is to test their build delivery system. While I am having a bit of a hard time finding the supporting quote, I did find one reference to it being the reason for ISOs being delayed.

And finally – we heard from you loud and clear you want ISO images of the new builds we release. We felt it was important to listen to that and give you what you want – but there’s a catch. Getting the update & install data from our Preview Builds mechanism is super important for us. It helps us ensure smooth ESD distribution, download, and upgrade success for this program going forward, and also will help us ensure great upgrades for people once we release Windows 10. So we’re going to release the ISOs at the same time as we publish to the Slow ring. That means if you want to be FIRST and FASTEST to get the build, you’ll need to use our Preview Builds mechanisms (either automatic or Check Now in PC Settings to download.) If you must have an ISO you’ll have to be a bit more patient. I hope that you’ll consider that a fair tradeoff.

So what is my point? Basically, it is difficult for us to make assumptions about how baked Windows 10 is from our standpoint. They are being more open with us than ever about their development methods, but we don't know certain key things. We don't know what final feature set they plan. We don't know how much work has been done on any individual feature since it was merged into a build that we saw. We also don't know how much has been done by third parties. In some cases, a release in three months could equate to like, six months of work for a specific team since their last contribution was merged. I do think that any major feature we see at BUILD will pretty much be the last additions to the OS before it launches though, unless they have a surprise that will surface at E3 or something.

Also, remember that the things they show us are slanted to what they want feedback about.

Source: Thurrott.com

AMD is making SeaMicro walk the plank

Subject: General Tech | April 20, 2015 - 01:17 PM |
Tagged: amd, seamicro, HPC

Just over three years ago AMD purchased SeaMicro for $334 million to give them a way to compete in HPC applications against Intel who had recently bought up QLogic and the InfiniBand interconnect technology.  The purchase of SeaMicro included their Freedom Fabric technology which was at that time able to create servers which could use Atom or Xeon chips in the same infrastructure.  AMD developed compatibility with their existing Opteron chips and it was thought that this would be a perfect platform to launch Seattle, their hybrid 64bit ARM chips on.  Unfortunately with the poor revenue that AMD has seen means that the SeaMicro server division is being cut so they can focus on their other products.  Lisa Su obviously has more information that we do on the performance of AMD but it seems counter-intuitive to shut down the only business segment to make positive income, but as The Register points out the $45m which they made is down almost 50% from this time last year.  AMD will keep the fabric patents but as of now we do not know if they are looking to sell their server business, license the patents or follow some other business plan.

seamicro_logo.png

"Tattered AMD says it's done with its SeaMicro server division, following a grim quarter that saw the ailing chipmaker weather losses beyond the expectations of even the gloomiest of Wall Street analysts."

Here is some more Tech News from around the web:

Tech Talk

 

Source: The Register

Podcast #345 - MSI GS30 Shadow, Gigabyte X99-SOC, Skylake Leaks and more!

Subject: General Tech | April 16, 2015 - 02:36 PM |
Tagged: x99-soc, video, Skylake, Samsung, podcast, nvidia, msi, motorola, Moto E, Intel, GTAV, gs30, gigabyte, Broadwell, amd, 840 evo

PC Perspective Podcast #345 - 04/16/2015

Join us this week as we discuss the MSI GS30 Shadow, Gigabyte X99-SOC, Skylake Leaks and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

The URL for the podcast is: http://pcper.com/podcast - Share with your friends!

  • iTunes - Subscribe to the podcast directly through the iTunes Store
  • RSS - Subscribe through your regular RSS reader
  • MP3 - Direct download link to the MP3 file

Hosts: Jeremy Hellstrom, Josh Walrath, Allyn Malventano, and Sebastian Peak

Subscribe to the PC Perspective YouTube Channel for more videos, reviews and podcasts!!

 

NVIDIA Released 350.12 WHQL and AMD Released 15.4 Beta for Grand Theft Auto V

Subject: Graphics Cards | April 14, 2015 - 01:27 AM |
Tagged: nvidia, amd, GTA5

Grand Theft Auto V launched today at around midnight GMT worldwide. This corresponded to 7PM EDT for those of us in North America. Well, add a little time for Steam to unlock the title and a bit longer for Rockstar to get enough servers online. One thing you did not need to wait for was new video card drivers. Both AMD and NVIDIA have day-one drivers that provide support.

rockstart-gtav_trevor2_1920x1080.jpg

You can get the NVIDIA drivers at their landing page

You can get the AMD drivers at their release notes

Personally, I ran the game for about a half hour on Windows 10 (Build 10049) with a GeForce GTX 670. Since these drivers are not for the pre-release operating system, I tried running it on 349.90 to see how it performed before upgrading. Surprisingly, it seems to be okay (apart from a tree that was flickering in and out of existence during a cut-scene). I would definitely update my drivers if they were available and supported, but I'm glad that it seems to be playable even on Windows 10.

Source: AMD

Podcast #344 - Intel SSD 750 Series, NZXT S340, an ASUS FreeSync Monitor and more!

Subject: General Tech | April 9, 2015 - 12:58 PM |
Tagged: video, podcast, nvidia, mg279q, Intel, gsync, gigabyte, freesync, ddr4-3400, corsair, compute stick, asus, amd, 750 series

PC Perspective Podcast #344 - 04/09/2015

Join us this week as we discuss the Intel SSD 750 Series, NZXT S340, an ASUS FreeSync Monitor and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

The URL for the podcast is: http://pcper.com/podcast - Share with your friends!

  • iTunes - Subscribe to the podcast directly through the iTunes Store
  • RSS - Subscribe through your regular RSS reader
  • MP3 - Direct download link to the MP3 file

Hosts:Ryan Shrout, Jeremy Hellstrom, Josh Walrath, and Allyn Malventano

Subscribe to the PC Perspective YouTube Channel for more videos, reviews and podcasts!!

AMD’s FX-8320E; get in the game for $138?

Subject: Processors | April 7, 2015 - 05:56 PM |
Tagged: amd, FX-8320e

Over at Techgage one of the writers recently updated their system, due to budget constraints they needed to stay in the $600-700 range all told which of course indicates an AMD build.  They chose the $138 FX-8320E for their processor, along with a pair of GTX 760s, the ASUS M5A99FX Pro R2.0, 8GB of DDR3-1866 and with storage, power, cooling and case they managed to keep within the ir budget.  The question remain is if it is powerful enough for reasonable gaming duties such as Borderlands 2.  Read on to see if the recommendation is to go with AMD or the i3-4330 and a low end H97 board.

AMD-FX-Processor-Box-500x535.jpg

"Released this past fall, AMD’s FX-8320E processor promises to deliver a lot of processing power for those on a budget. It sports eight cores, and as a Black Edition, its overclocking capabilities are unrestricted. But is that enough to make this the best go-to budget processor, especially for gamers?"

Here are some more Processor articles from around the web:

Processors

Source: Techgage