AMD's Ryzen CPUs (Ryzen/TR/Epyc) & Vega/Polaris/Navi GPUs

Not if AMD chose to simply move the tiers of CPU's... in which case, they could use existing Threadripper as a baseline and move it to 12nmLP with higher clocks - at which point, you end up with higher core count - but TR is rather big in comparison to regular Ryzen, and not sure if a shrink to 12nm LP would be enough to bring it down to regular socket size... even with 15% higher density (I think 7nm would be needed for that).

As I already said, it's unlikely that this is accurate and we are more likely to see existing core counts (like on original Ryzen) with those clock increases.

 

AMD Issues Official Statement Regarding RX 560 Silent Downgrade
https://www.techpowerup.com/239486/amd-issues-official-statement-regarding-rx-560-silent-downgrade

Spoiler: Details...

"AMD has come forward with a statement regarding the state of things as it purports to the recent RX 560 downgrade issue, which has been covered extensively by media outlets just this week. The issue stems from the fact that there was no clear differentiator between two different versions of AMD's RX 560 graphics card, which could ship with two different levels of performance: one with 14 CUs (Compute Units) enabled (896 stream processors) or 16 (1024 stream processors, the original specification for the card).

"It's correct that 14 Compute Unit (896 stream processors) and 16 Compute Unit (1024 stream processor) versions of the Radeon RX 560 are available," stated a company representative. "We introduced the 14 CU version this summer to provide AIBs and the market with more RX 500 series options. It's come to our attention that on certain AIB and e-tail websites there's no clear delineation between the two variants. We're taking immediate steps to remedy this: we're working with all AIB and channel partners to make sure the product descriptions and names clarify the CU count, so that gamers and consumers know exactly what they're buying. We apologize for the confusion this may have caused."

While it's true that AMD is now dealing with the situation openly, the company itself isn't completely blameless on the issue. AMD's own RX 560 product page lists the two possible configurations (14 and 16 CU) absent of any further differentiation in product name. At the same time, AMD could have better enforced differentiation opportunities with its AIB partners upon supplying them with both versions of the GPU.

As it stands, some users have been unwittingly buying a lower performing part than what was initially advertised by AMD. While this new transparent stance from AMD is refreshing, it doesn't do much regarding already affected customers. Of course, if users suspect or want to check whether or not they have purchased the higher performance version of the card (16 CUs), they can perform a GPU verification with TechPowerUp's GPU-Z tool."

AMD Apologizes For RX 560 Spec Change, Ensures Full Transparency With AIB Partners
http://www.tomshardware.com/news/amd-radeon-rx560-spec-change,36061.html

"AMD issued a new statement regarding its recent spec change to its Radeon RX 560 graphics card.

The company recently came under fire for a sneaky specification change to the RX 560, with the company now offering two different versions of the same product. Now, there are RX 560s with 14CUs (896 stream processors) and some with 16CUs (1,024 stream processors, the original specification). This sent the internet into a rage of sorts, with AMD initially passing the buck in regards to disclosure, stating that AIB partners would be responsible for being transparent with its individual RX 560 offerings. Now, the company appears to have changed course and issued a new statement:

“It’s correct that 14 Compute Unit (896 stream processors) and 16 Compute Unit (1024 stream processor) versions of the Radeon RX 560 are available,” stated a company representative. “We introduced the 14CU version this summer to provide AIBs and the market with more RX 500 series options. It’s come to our attention that on certain AIB and e-tail websites there’s no clear delineation between the two variants. We’re taking immediate steps to remedy this: we’re working with all AIB and channel partners to make sure the product descriptions and names clarify the CU count, so that gamers and consumers know exactly what they’re buying. We apologize for the confusion this may have caused.”

Although this statement indicates that the company is now dealing with the potential (and probable) confusion caused by two products with different performance sharing the same name, it’s interesting to note that AMD made this change over the summer, indicating that anyone who purchased an RX 560 graphics card between now and then may have purchased the 14CU version of the card without knowing it.

Consumers are understandably furious over the secretive change, with many commenters in our forums denouncing the move as “Nvidia-like” and comparing it to the GTX 970 memory segregation conundrum from a few years back. However, whereas the GTX 970 situation didn’t affect bottom-line performance of the GPU, AMD’s specification change portends a significant difference in performance between two products with the same name.

For anyone who purchased a Radeon RX 560 graphics card in the past few months, a quick look at GPU-Z will give you the stream processor count for your GPU. However, if you did happen to unknowingly purchase a 14CU version of the RX 560, there’s currently no set means of restitution. We reached out to AMD about this, and will update the article accordingly with the company’s response."

 

AMD To Stop Shipping Reference RX Vega Cards In Favor Of Custom Models
https://wccftech.com/amd-stops-shipping-reference-rx-vega-favor-custom-cards/

"AMD is discontinuing the production and distribution of reference Radeon RX Vega 64 and 56 graphics cards,reports computerbase.de. All major retailers in Germany have stopped carrying the reference designed versions of both cards, in favor of custom AIB models.

AMD Stops Shipping Reference RX Vega 64 & 56 Cards In Favor Of Custom Models
We have yet to see this take place among US e-tailers. In fact, neither does Amazon or Newegg appear to have any custom AIB versions of neither RX Vega in stock, as both e-tailers continue to only carry the reference RX Vega 64 and 56 designs. Although this is expected to change over the coming weeks, as more AIBs ship custom models and as the existing inventory of reference cards runs out.

ComputerBase cites several sources and reports that AMD has already stopped shipping reference Vegas in favor of allocating more Vega 10 GPUs for the company’s add-in-board partners to build custom designs with. Asus, Gigabyte, Sapphire, PowerColor & XFX have all debuted custom versions of AMD’s RX Vega 64 and 56 graphics cards, the majority of which are already shipping to retailers world-wide, including stores like Best Buy and Microcenter.

It’s not yet clear whether AMD will resume production of any specific RX Vega design at a later date. With 2017 coming to a close very soon, the company is no doubt already busy with the new 12nm Ryzen & Vega chips expected in the first half of next year.

Refreshed Vega parts are expected to include the all new Vega 11 mid-range GPU replacing Polaris 10/20 and the RX 580/70 series. Additionally, a faster and more power efficient iteration of the Vega 10 GPU i.e. Vega 20 could also be in the works to debut some time in the first half of next year.

These refreshed parts will have to hold the line against NVIDIA’s upcoming Volta cards, which are expected to debut in the first half of next year. That is until Navi makes landfall, which is believed to be happening some time in the second half of 2018 around August. All in all, 2018 is shaping up to be a pretty exciting year for graphics with Volta, 12nm Vega, 7nm Navi all making their debut in one form or another."

 

As for the 5GHz with Ryzen 2 that seems about right. Ryzen 2 is supposed to be 7nm and also should be capable of a redo making it a 12 core chip. while I agree it is just a rumor to me it seems viable as well. Love to eventually see a new TR at 24C, 48T and well over 5 GHz. as that would be a worthwhile upgrade.

 

Another website with the rumour:
https://hothardware.com/news/amd-zen-2-leak-12-core-ryzen-7-2800x-51ghz-boost

They are calling Ryzen refresh as Ryzen 2 - which is stupid. Wouldn't it be easier to just call it Ryzen+ to signify a refresh and Ryzen 2 (for 7nm) and Ryzen 3 for when they arrive?

As for the 2800x (refresh) boosting to 5.1 GhZ... bear in mind that the 7nm process would allow the BASE clocks to be at 5GhZ... the apparent 'leak' indicates turbo boosting to 5.1GhZ (which is obviously not base speed).

The leak however may be at least partially accurate in the sense of clock speeds being as presented... but not necessarily the core counts.
As I said before, AMD didn't mention anything about increasing core count on the refresh... but, the new process DOES allow for 15% higher chip density... I just don't know if this is enough to include that many extra cores and fit it all in the same (smaller) Ryzen socket since 2nm decrease is not exactly a lot.

The main advantage of the process are higher clock speeds because it's designed for high performing parts... so, it stands to reason the base clock increases would go beyond 10% (with bare minimum increase being 10%)... more in line with 20% increase on the base (considering the process AMD is transitioning from).

At any rate... we need more information.

 

After that VEGA marketing and Bulldozer marketing I don't take words about 5GHz+ OCing on fate Maybe in some conditions some lucky owners - sure. But every CPU easily reaching 5GHz - nope, I don't buy it unless I see it.

 

Higher clocks(Turbo boost) yees, but not on all cores. And the OC headroom will decrease a lot. Don't ecxpect magic. Aka Fully package with +5GHz all cores + nice oc headroom on top.

 

It's 12nm Leading process designed for high perfoming parts.
14nm was for low performing parts (and low clocks).
This new process should allow Ryzen to catch up to Intel in clocks and probably surpass it in some areas.
If AMD also includes higher core counts in the refresh with these high clocks, then all the better. You get A LOT more performance for your buck at any rate... and higher performing parts than Intel in most scenarios. Even single core performance on a 1700 replacement will catch up to base Intel for the most part.
Again, I think the x parts might be better binned so they could allow for better overclocking - but I don't think anyone said anywhere that ALL these parts will be reaching 5GHZ on overclock unless they came like that from factory.
5GhZ base however is for 7nm (Ryzen 2).

As for Vega... its marketing was a bit off, yes, but AMD was primarily messed over by the manuf. process they used... and overvoltage for power consumption.
On another note, Vega 56 and 64 perform really well in PC optimized games, along with those that have DX12 and Vulkan API's.
The Vega64 2k and 4k performance is within 10% of 1080ti... which is pretty good if you ask me... whereas Vega 56 at 2k and 4k is at/above 1080 level.

Undervolt them, and you increase their performance by another 5% by removing the thermal headroom and drop their power consumption to Pascal levels.

Vega refresh on 12nm LP should bring about gaming optimised Infinity Fabric (if Raja was not talking fibs), and far better/higher clocks in both core and HBM at a lower power draw (thanks to the far superior manuf. process).
Alas, the process is far better, but we have to see how Glofo handles the yields on it... they had an issue with GPU yields before, but that also could have been due to working with a relatively unsuitable manuf. process.
We have to wait and see again.

 

Fake AMD Ryzen 2800X 12 Core 5.1GHz Slide Sends Media Into Frenzy

"The optimists among you will want to believe that this is true. After all who wouldn’t want to see 12 core Ryzen processors at as low as $329 and with clock speeds as high as 5.1Ghz, coming out on the 1st of February no less. The skeptics among you will point out that the specifications, pricing and timeline are all too conveniently fantastical. Sadly, the rumor has spread like wildfire with no evidence, other than the slide above, to corroborate any parts of it."

"As a matter of policy we ignore any unsubstantiated rumors or leaks that we’re unable to corroborate through our secondary sources. This time however, we decided to make an exception after seeing how quickly and wildly this fake slide has spread."

 

Considering that this is 12nm LP, I think we could see better overclocking headroom than Intel's current parts, but yes, boosting on all cores to much higher clocks than original Ryzen is pretty much guaranteed.
If this process is IBM based... I wonder, is IBM's manuf. process better or worse than Intel's?

5GhZ boost across all cores... not so much. 4.8 GhZ maybe if AMD sets it like that with single core going to 5.1GhZ on 2800x.
Manual OC-ing should be able to push all cores on 5GhZ though.

For TR and EPYC... I think we can at least apply 20% boost on single and multi-cores in comparison to existing TR and EPYC.

 

That's why we're being cautious.
The 12nm Leading process is certainly a lot better/superior to one Ryzen is currently on, so at least the clocks themselves could be accurate, while we could/will likely get the same core count.

 

High clocks see INTEL means less cores vs. fixed TDP(a reason Intel went for 6 cores). You can't get both... 8/10 or 12 cores and hope on very high clocks and within 90/95w TDP for mainstream chips. Not early as middle 2018. Too early.

 

Well, we do know that AMD plans on sticking to the same TDP as Ryzen 1... but the 12nm LP includes 15% higher density, and as I said before, not sure if that's enough to provide more cores, which is why its more likely to see those core speeds as provided by the leaks, but not the core increase.

Yes I am aware that for Intel, high clocks mean less cores... but this new process is smaller (and possibly better) than Intel's (since it's 12nmLP vs 14nm++), and AMD doesn't necessarily have to conform to the same premise.
AMD's Raven Ridge for example can boost to 2.2GhZ across all cores, whereas Intel's 8250u can't.

This new process also allows lower voltages... so that means lower power consumption, better efficiency.
Besides, 5.1 GhZ is boost on a single core.
If current Ryzen 1800x can boost to at least 4Ghz across all 8 cores... a 20% boost would result in 4.8GhZ for single core, and about 4.4GhZ for all 8 cores.

Alas, since this leak is basically a rumour... we can only speculate. We need verification and more information.
Plus, I don't expect to see Ryzen+ so early in 2018... it may be possible, but that depends if 12nmLP is ready or not.

 

AMD have to launch new cpu models early 2018 <if> they don't have a lot faster cpu's. This due hope for high sales before Intel start push out their own. If they wait, they will come to see higher competition from INTEL and need fastest possible chips. The winner then will be those who can offer the fastest chips. Of course price matters, but for some is faster chips the only way to go. And yeah, We need verification and more information. Rumors ain't the real truth

 

My bad.
I checked over the info on 12nm LP and it seems that AMD's papermaster said the process will be ready for February 2018.

This now leaves the question of, how much faster will Intel's next architecture be in comparison to their existing one?

Also, if AMD releases Ryzen 2 (7nm) in late 2018, Intel might find itself in a sort of a 'grey area' in between AMD's releases.

 

Fake AMD Ryzen 2800X 12 Core 5.1GHz Slide Sends Media Into Frenzy
https://wccftech.com/fake-amd-ryzen-2800x-12-core-5-1ghz-slide-sends-media-frenzy/

"A purported “leak” has been making the rounds on the tech news media showing three alleged upcoming Ryzen 7 2000 series CPUs, Including Ryzen 7 2800X, 2800 and 2700. The slide in question is surprisingly detailed, outlining specifics such as core & thread counts, clock speeds, prices and launch dates.

Spoiler: Details...

The optimists among you will want to believe that this is true. After all who wouldn’t want to see 12 core Ryzen processors at as low as $329 and with clock speeds as high as 5.1Ghz, coming out on the 1st of February no less. The skeptics among you will point out that the specifications, pricing and timeline are all too conveniently fantastical. Sadly, the rumor has spread like wildfire with no evidence, other than the slide above, to corroborate any parts of it.

What’s more, none of the news media reports surrounding this slide that we’ve come across include any references to its original source. And unfortunately, seemingly no effort has been put forward to identify the original source that had allegedly leaked it.

As a matter of policy we ignore any unsubstantiated rumors or leaks that we’re unable to corroborate through our secondary sources. This time however, we decided to make an exception after seeing how quickly and wildly this fake slide has spread.

After seeing it for the first time it immediately jumped at us as a fake for obvious reasons. But we’ve also tried our best to do our due diligence and ran it through the usual channels which have so far been coming out in agreement that it is indeed fake. We took it a step further and analyzed the slide forensically. Apart from dead giveaways like the fact that the font for some of the characters displayed in the slide is different from what AMD uses in its own slides we’ve also found evidence that indicates the image has been visually manipulated/edited.

We have also attempted to track the original source of the image down and after some effort we believe we have successfully done so. The image is more than a month old, and we believe that it made its way through the web via 4chan. We also believe that it was meticulously crafted to intentionally mislead and fool, rather than exist simply as a product of hopeful conjecture or a joke by some random enthusiast on some tech forum."

But Wait, There’s More! – 5.5GHz 24 Core Threadrippers For Everyone? Or Maybe NOT…


"And as if 12 core mainstream Ryzen 7s weren’t enough we also found a couple of other silly… I’m sorry “leaked” slides allegedly showing AMD’s upcoming 3000 series Threadripper CPUs. Chips with as many as 24 cores and 48 threads and clock speeds as high as 5.5GHz…

Yes folks, we’re apparently supposed to believe that AMD’s new Threadripper lineup is comprised of diamond melting products. The same source also claims that the Ryzen 2000 series will be built on 7nm technology which we also know to be completely false.

If you want to know when AMD’s upcoming Ryzen 2000 series lineup will be out and what the new parts will actually bring to the table, yesterday’s article has you covered. In the meantime, try to take all rumors with the usual grain of salt."

 

Really obnoxious when people do this.

 

I think the 12nm is slated for Q1 2018. The 2000 series is supposed to be 7nm if I am not mistaken. now I could see the 2000 series being Q1 2019 and maybe the TR 3900's being Q3 2018, AMD is great for delays though. They could have been slide designed for release in 2018 at some point, again this is all rumor and speculation.

I can say if that 3970x comes to market I would grab it up right away.

 

I still maintain there's a chance we could see those baseline clock speeds across all cores (but not 5Ghz and above) due to the new process, without an increase in core count.

 

At 12nm I doubt it will have a core increase. At 7nm I can easily see that happening. 12nm should be just a better overall process than 14nm and hopefully better clocks and slightly lower vcore.

 

Most definitely much higher clocks on 12nm (up to or even beyond Intel level if the process is up to snuff and it does seem to be considering even TSMC will be using it).
As for core increases, as I said, I'm sceptical of that too... but the process also offers 15% higher chip density, so I'm wondering what AMD will do to fill that space up.
More transistors probably, but what exactly will they do?

 

Remotely possible but I'm not holding my breath.

 

AMD Chipset Drivers Download v17.40
"Ready for AMD Ryzen processors - AMD Socket AM4 represents the company’s new future-proof platform targeting the fastest DDR4 memory, PCIe 3.0, and NVMe technologies available, as well as the first native USB 3.1 Gen2 support on a chipset. With processor-direct SATA and USB connectivity, configurable for real-world flexibility, the new AM4 platform takes advantage of the leading-edge features of today, and tomorrow."

This driver supports the X370, B350, A320 and A300 chipsets. Supports Windows 7 and 10.

 

Once I get my GL702ZC laptop back and more money, I'll definitely get the higher clocked RAM most likely, or at least one stick to test out the RAM frequencies.

Asus better release a BIOS update to support Ryzen+, Ryzen 2 and Ryzen 3 though.
I'd like to upgrade the CPU eventually to Ryzen 2 most likely on 7nm and equal to or lower than 65W TDP.

 

Some other news on a possible desktop APU with Vega 28:
https://pcgamesn.com/amd-ryzen-mobile-gaming

Interesting.
If this thing is real and has a TDP of 65W, they can easily cram it inside a nice 15" laptop with proper cooling.
The GL702ZC has a CPU and GPU each rated at 65W (so, technically 130W TDP)... and this APU would be HALF that while providing nearly identical graphics performance.

Possible speculation if this APU is real for 7nm: Ryzen 2 8c/16th with Navi igp and Vega 56 level performance or higher?

 

Unless that 28 CU GPU is clocked very high (unlikely based on the TDP) it won't have "nearly identical graphics performance" to a 36 CU GPU such as the RX 580.

 

30% difference in CU's... still, you have to keep in mind that when Vega 56 and 64 were tested on identical clocks in games, the differential in their stream processors didn't manifest in performance (except that Vega 56 was consuming less power due to having less stream processors).
In other words, both were performing identically in games on identical clocks.

Also bear in mind that RX 580 (desktop) was downclocked by 20% from the desktop version and its TDP brought down to 65W (limited to 68W) for the mobile version (and is built on Glofo 14nm manuf. process designed for low clocks and low power).

This APU is likely built on 12nmLP which is designed for high clocks, and has unknown VRAM to boot (possibly HBM, but we won't know this until more details emerge if the APU is real).
Vega however is generally suffering from bandwidth starvation... which is alleviated via HBM overclocking... so, if the core clocks are relatively low but HBM (if it has it) frequencies high, you could end up with comparable performance.

Plus, what's the IPC difference (if any) between Vega and Polaris as far as architectures go exactly?

 

AMD Ryzen™ Threadripper™ for 3D Artists – Mike Pavlovich

 

Nice thing about Ryzen, and TR, I that you can build a system and with the savings get a 1080 TI over a 1080 for better gaming systems on the same budget. Once Ryzen is in the mobile space we will see better market share for AMD systems.

 

has anyone else noticed since the last W10 update that performance of the AMD has severely suffered? I notice this on single threads and even CB R15 has lost 50 points or so. SuperPI single thread has gone on 16k from 0.140 to over 0.350 all of a sudden and it is on power performance profile.

Edit; definitely something funky. CPU-z shows 4023 MHz but where task manager used to be fixed is now varies under load from idle at 3.88 GHz to 3.96 or so. This is under power or Ryzen balanced profile. To be sure I went with the older bios 1.7 for my board and it is the same as the newer one.

 

The windows update might not be playing well with the CPU overclock.
Though the loss seems... minimal.
Still, it might be worth noting this to Microsoft. I'd hate for them to be crippling AMD intentionally or are primarily releasing updates verified for Intel, but AMD comes second.

 

I don't play with Windows 10, but it seems MS "fixed" the power plan UI with some simple slider that doesn't actually give full performance at top slide, so you have to dig into the settings (advanced), or find the original interface UI - I've heard it's there somewhere, and revert back to using that.

Makes sense that if your power plan isn't maximum CPU 100%/100%, it's not gonna have the same full performance.

 

It is weird, I got back most of the CB r15 performance by eliminating a lot of the startup. Not the single thread though. I just installed Linux Mint 18.2, tried 18.3 but it kept failing. I may just have to give up on W10 except when needed.

 

I should note M$ seems to fix whatever was hanging W10 up. Still have the strange CPU speed reading issue in Task Manager.

 

I think the CPU frequency reading in Task Manager is just bugged. I have speedstep disabled in the BIOS, and I know my CPU is fixed at 4.7Ghz (albeit with C-states active) using Windows High Performance Power Profile, and Windows Task Manager is showing 4.0 - 4.7Ghz fluctuating, whereas all other measuring solutions are showing 4.7Ghz fixed.

 

Clock speed in Win task manager have never worked the way it should. Put in two different clocks and you will get two different readings although you changed only one bin.

 

I spoke too soon. It seems that the single thread slowdown is back. W10 just screws up everything..

 

Tested Fully updated Creators update vs. Creators Fall update?

 

I am not sure what is hanging it up. Linux mint has no issues. I could tell at fist there was an issue as the OS just lost its snappiness. I suspected and found that a single non optimized thread slowed way down. I verified it by testing with SuperPI MOD, and the times went from 0.140 sec yo do 16k to 0.350 sec's.

I posted before as the times went back to 0.142 Sec's. Now though it is back to 0.350 sec's again.

 

Interesting, so the performance results are varying depending on when you run them? I wonder what the correlating item(s) are?

Processes running to a schedule, random start up of services, or what?

Maybe it will take a bit more instrumentation to see what's running at the same time?

 

Look at passmark at install and then now;

 

And, you have found that huge drop in performance is due completely to the Windows Update?

 

I had noticed since, I have not found the actual reason but there are no other changes. CB R15 numbers though are fine so it seems optimized multithread is not affected.

 

I'd do what Papusan suggested there. Use a backup to take yourself back to Creators Update & then retest your Passmark, then go back to the Fall Creators Update & test your Passmark - see if there's a difference. (If you've got a Macrium Reflect saved disk image, then it's just a matter of minutes to restore to a previous image - takes me 5 mins). (check that you've not installed any programs or anything since that could be different & affecting your results between the two images).

 

Try this tweak below after you have forced Win maintenance. Let it be disabled a couple of days. Try again.

Disable Automatic Maintenance in Windows 10. Test without this trash fireing up. I disable it with tasks. If I want it running, I just enable it and then force Maintenance. Then disable the task again. No need for this trash running several times, days/week. M$ Morons want to see what's in your computer. Even less needed the whole time if you prefer prevent what Windoze will determine what you should download. Windoze X is a BIG SPYBOX!!

 

I did a system reset and SuperPI scores came back Now I need for the install to settle so I can test passmark.

Edit; have to reinstall video drivers but CPU is now 25,796. Much better. In the end this stinks having to redo the install.

 

Try cleaning up superseded SxS file via DISM and compile ngen assembly files optimised for your system to bring down the CPU usage.

 

AMD Promontory 400-chipset spotted at PCI-SIG
https://videocardz.com/74528/amd-promontory-400-chipset-spotted-at-pci-sig

"AMD Ryzen 2000 AMD 400-series motherboards are coming

The refresh of the first-generation Ryzen is currently expected in the first quarter next year. Although new CPUs will certainly be compatible with 300-series motherboards, AMD is giving a chance to their partners to refresh their motherboard offering with X470, B450, A420 series.

The Zen+/Ryzen 2000, also known as Pinnacle Ridge, is a 12nm upgrade of Zen1. We expect higher clock speeds and better power efficiency. Meanwhile, Zen2, which might be used for Ryzen 3000 series, is said to deliver bigger chances to Zen architecture as we approach 7nm fabrication process.

The Promontory platform, which is a codename for Summit Ridge 300 series chipsets, has been listed at PCI-SIG since January 2016. However, for the first time, it was also listed as a codename for ‘400-Series Chipsets’."

Source: PCI-SIG

 

Just as a counterpoint; all systems I've used (all Intel) after the latest update are noticeably more responsive.

Wonder what AMD/MS is doing to step on each other's toes?

 

I am not sure what it was. I can say in researching this I found quite a few people complaining and it was not just AMD people either. Just glad to have my desktop back on steroids.