Amd Ryzen Threadripper & X399 MegaThread! FAQ & Resources

[8350rocks]

I did read it and that is why I know the article tries to rewrite the history and omits to mention the SP4 platform and the corersponding 8/12/16 core CPUs.

It omits to mention any platforms or codenames. Zen is not a platform - at least not in the way it's said in the article.

This is Forbes, not Hexus, Anand or SemiAccurate. idk why Forbes would deviate from a business/human centered story on TR to talk about an irrelevant scrapped product from an irrelevant scrapped product line. You expect them to talk about Zeppelin too?

The Forbes article claims (bold emphasis mine)

It might surprise you to learn that Ryzen Threadripper wasn’t originally part of AMD’s plans, which back in 2015 only included Ryzen and the EPYC server CPU.

The bolded part is not true, because RyZen is AM4 platform, and EPYC is SP3 platform and AMD also planned Snowy Owl which used the SP4 platform. The problem is not Forbes but the AMD people giving the interview and trying to rewrite the history to hide the SP4 platform, which appeared in roadmaps one or two years ago but now is silently gone and AMD no longer talks about it.

http://wccftech.com/amd-cpu-roadmap-leak-7-nm-starship-14nm-naples-snowy-owl-zen-core/ ?

I can't find any slides from before that.

Just because it was leaked in 2017 doesn't mean anything. Pay attention to the slides.

Instead of being a jerk, you could provide a link at least? And I did look at the slides. They mention the sockets and that is a product paired with Naples at the top. Are you absolutely sure you're not just confusing the codenames with the actual product names that they were released under?

A link to what?

What confusion? It is all well-established in AMD internal docs

Ironically all of the ZP platforms in the bit you are referencing are all extinct aside from SP3...it would appear those were engineering platforms, nothing more. RV still shows AM4.

 

[juanrga]

I did read it and that is why I know the article tries to rewrite the history and omits to mention the SP4 platform and the corersponding 8/12/16 core CPUs.

It omits to mention any platforms or codenames. Zen is not a platform - at least not in the way it's said in the article.

This is Forbes, not Hexus, Anand or SemiAccurate. idk why Forbes would deviate from a business/human centered story on TR to talk about an irrelevant scrapped product from an irrelevant scrapped product line. You expect them to talk about Zeppelin too?

The Forbes article claims (bold emphasis mine)

It might surprise you to learn that Ryzen Threadripper wasn’t originally part of AMD’s plans, which back in 2015 only included Ryzen and the EPYC server CPU.

The bolded part is not true, because RyZen is AM4 platform, and EPYC is SP3 platform and AMD also planned Snowy Owl which used the SP4 platform. The problem is not Forbes but the AMD people giving the interview and trying to rewrite the history to hide the SP4 platform, which appeared in roadmaps one or two years ago but now is silently gone and AMD no longer talks about it.

http://wccftech.com/amd-cpu-roadmap-leak-7-nm-starship-14nm-naples-snowy-owl-zen-core/ ?

I can't find any slides from before that.

Just because it was leaked in 2017 doesn't mean anything. Pay attention to the slides.

Instead of being a jerk, you could provide a link at least? And I did look at the slides. They mention the sockets and that is a product paired with Naples at the top. Are you absolutely sure you're not just confusing the codenames with the actual product names that they were released under?

A link to what?

What confusion? It is all well-established in AMD internal docs

Ironically all of the ZP platforms in the bit you are referencing are all extinct aside from SP3...it would appear those were engineering platforms, nothing more. RV still shows AM4.

The only "extinct" platforms are those based in SP4... and that was precisely my point. AMD marketing team through that Forbes article pretends that the SP4 platform didn't exist, when some of us know the history begin them.

 

[Shotta06]

Juan, you tend to be mis-informed majority of the time I see you post. Is what your stating as facts something 100% accurate.. or something you think know

 

[-Fran-]

I can correlate because I know the information. I have known it for while. Even before leaks, as the one that you linked to.

No, my thesis is not "TR = Snowy Owl". The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC.

I don't think anyone understands you Juan, but it's fine.

My friends and colleagues say otherwise

And doesn't that statement validate AMD's story?

So, I devote some time to explain that the Forbes article is a piece of marketing that tries to rewrite the history and you interpret my words as I am validating AMD history. ROFL!

If you're saying the current platform for TR is a merge of two other platforms that were not going to see the light of day, then how do you take that? I'm not sure if you're being obtuse on purpose or just want to nitpick something that is really not important to nitpick.

Cheers!

 

[juanrga]

I can correlate because I know the information. I have known it for while. Even before leaks, as the one that you linked to.

No, my thesis is not "TR = Snowy Owl". The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC.

I don't think anyone understands you Juan, but it's fine.

My friends and colleagues say otherwise

And doesn't that statement validate AMD's story?

So, I devote some time to explain that the Forbes article is a piece of marketing that tries to rewrite the history and you interpret my words as I am validating AMD history. ROFL!

If you're saying the current platform for TR is a merge of two other platforms that were not going to see the light of day, then how do you take that? I'm not sure if you're being obtuse on purpose or just want to nitpick something that is really not important to nitpick.

No. I am not saying that. READ WHAT I WROTE:

The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC.

Did you read it? Ok, then where according to you I am saying that "TR is a merge of two other platforms that were not going to see the light of day"? I am stating TR is a merge of SP4 and SP3 and I am saying the SP3 platform saw the light as EPYC.

It is ok if you cannot read plain English but stop insulting me.

 

[-Fran-]

I can correlate because I know the information. I have known it for while. Even before leaks, as the one that you linked to.

No, my thesis is not "TR = Snowy Owl". The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC.

I don't think anyone understands you Juan, but it's fine.

My friends and colleagues say otherwise

And doesn't that statement validate AMD's story?

So, I devote some time to explain that the Forbes article is a piece of marketing that tries to rewrite the history and you interpret my words as I am validating AMD history. ROFL!

If you're saying the current platform for TR is a merge of two other platforms that were not going to see the light of day, then how do you take that? I'm not sure if you're being obtuse on purpose or just want to nitpick something that is really not important to nitpick.

No. I am not saying that. I didn't say that "two other platforms that were not going to see the light of day". READ WHAT I WROTE. I said it is a merge of SP4 and SP3, and SP3 saw the light... It is EPYC, as I have said a dozen of times.

Nitpicking again...

So, you're saying that something that did not exist before (in your own words) became something after 2 *different* things came together (again, by your own words), can't be something new, because AMD planned for it and displayed it in their road-maps as the 2 separate things.

Am I reading you correctly there? I don't mind you drawing pictures to explain your point, to be honest, since you're not making sense here.

 

[juanrga]

I can correlate because I know the information. I have known it for while. Even before leaks, as the one that you linked to.

No, my thesis is not "TR = Snowy Owl". The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC.

I don't think anyone understands you Juan, but it's fine.

My friends and colleagues say otherwise

And doesn't that statement validate AMD's story?

So, I devote some time to explain that the Forbes article is a piece of marketing that tries to rewrite the history and you interpret my words as I am validating AMD history. ROFL!

If you're saying the current platform for TR is a merge of two other platforms that were not going to see the light of day, then how do you take that? I'm not sure if you're being obtuse on purpose or just want to nitpick something that is really not important to nitpick.

No. I am not saying that. I didn't say that "two other platforms that were not going to see the light of day". READ WHAT I WROTE. I said it is a merge of SP4 and SP3, and SP3 saw the light... It is EPYC, as I have said a dozen of times.

Nitpicking again...

So, you're saying that something that did not exist before (in your own words) became something after 2 *different* things came together (again, by your own words), can't be something new, because AMD planned for it and displayed it in their road-maps as the 2 separate things.

Am I reading you correctly there? I don't mind you drawing pictures to explain your point, to be honest, since you're not making sense here.

Your interpretation of my words goes in the opposite direction to what I really said.

I say that TR and Snowy Owl are different and you interpret that I said "TR = Snowy Owl".

I say that TR4 is a merge of SP3 (EPYC) and SP4 (Snowy Owl) and that SP4 is gone and you interpret that I said both platforms are gone: "TR is a merge of two other platforms that were not going to see the light of day".

And so on...

I stop here. Interpret my words as you want. There is a small but non-zero statistical chance you will get it.

 

[-Fran-]

I can correlate because I know the information. I have known it for while. Even before leaks, as the one that you linked to.

No, my thesis is not "TR = Snowy Owl". The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC.

I don't think anyone understands you Juan, but it's fine.

My friends and colleagues say otherwise

And doesn't that statement validate AMD's story?

So, I devote some time to explain that the Forbes article is a piece of marketing that tries to rewrite the history and you interpret my words as I am validating AMD history. ROFL!

If you're saying the current platform for TR is a merge of two other platforms that were not going to see the light of day, then how do you take that? I'm not sure if you're being obtuse on purpose or just want to nitpick something that is really not important to nitpick.

No. I am not saying that. I didn't say that "two other platforms that were not going to see the light of day". READ WHAT I WROTE. I said it is a merge of SP4 and SP3, and SP3 saw the light... It is EPYC, as I have said a dozen of times.

Nitpicking again...

So, you're saying that something that did not exist before (in your own words) became something after 2 *different* things came together (again, by your own words), can't be something new, because AMD planned for it and displayed it in their road-maps as the 2 separate things.

Am I reading you correctly there? I don't mind you drawing pictures to explain your point, to be honest, since you're not making sense here.

I am gone from trying to explain you again what I said. Please don't attribute me stuff not only I didn't say, but goes in the opposite direction to what I really said..

"The TR4 socket used in ThreadRipper is a fusion of the old SP4 socket used in Snowy Owl and the SP3 socket used in EPYC."

How can that be read otherwise? HOW?!

 

[randomizer]

I suggest dropping this topic before it goes any further downhill.

 

[juanrga]

I suggest dropping this topic before it goes any further downhill.

Agree! I stopped >>here<<

 

[goldstone77]

An AMD Threadripper X399 Motherboard Overview: A Quick Look at Seven Products
by Ian Cutress & Joe Shields on September 15, 2017 9:00 AM EST

With the release of AMD’s Threadripper CPUs into the HEDT market, board partners have released new motherboards based on the X399 chipset. Consumers are going to see quad channel memory, native 4-Way SLI and Crossfire capabilities, more full-speed M.2 slots, added 10G network ports, and more on the new platform. We're taking a quick look at each of the motherboards that the vendors are promoting in the market, as well as a few upcoming teasers.
AMD Ryzen Threadripper Needs a Mother(board)
A big takeaway from our AMD Ryzen Threadripper 1950X and 1950X review is the increase in CPU connected PCIe 3.0 lanes. A full 60 lanes are available to the PCIe slots and M.2 drives, while 4 lanes head off to the chipset where it's muxed for USB, SATA, low-level I/O, and other features. The increase in CPU PCIe lanes is set up to allow for a lot more flexibility connecting devices using PCIe bandwidth.

The four major players, ASRock, ASUS, GIGABYTE, and MSI, have launched X399 motherboards for Threadripper. These boards equip the massive 4094-pin socket with eight DIMM slots to cover the four memory channels and then differentiate themselves in PCIe configuration, IO, styling and software support.

Unlike Ryzen, the base processor is not a true SoC as the term has evolved over the years. In order to get the compliment of SATA and USB ports, each Threadripper CPU needs to be paired with an X399 chipset. So aside from the CPU PCIe lanes, the 'new' X399 chipset also gets some IO to play with:

Unlike mainstream platforms that have several chipsets to choose from that vary in capability and cost, the AMD high-end platform only has a single chipset, similar to Intel's X299. As we've learned over the last few months, some elements of the motherboard base cost, such as the socket, can cost a pretty penny to the board manufacturers, which in turn will be forwarded on to the user. These motherboards appear to start somewhere around $249-$299, with the top board so far set at $549. Guess which one AMD sent us for our CPU review.

The Underlying Block Diagram
When PCIe is the game, it becomes time to play with controllers and ports to see what fits. It all starts with the underlying diagram for Threadripper and X399:

AMD's marketing on the left is set out to explain that each product in the Threadripper and X399 line will be 'fully enabled' - none of the chips will be limited in PCIe lane counts or DRAM for the sake of segmentation. AMD is happy at this point to keep segmentation limited to cores, frequency, and L3 cache.

The diagram shows the two hubs for everything to connect to: the CPU and the chipset. The CPU has 60 PCIe lanes, provided in 3 or 4 PCIe root complexes at x16 each (the diagram above suggests x16/x8/x16/x8). These will often go towards the main PCIe slots to support add-in cards like GPUs, or they could be partitioned off for RAID controllers or fast ethernet if required. In total, AMD has stated that the platform can support six GPUs maximum. The CPU also supports eight USB 3.1 (5 Gbps) Type-A ports, quad channel DDR4, and support for up to three PCIe 3.0 x4 storage drives for 3-way NVMe.

Threadripper CPUs reserve four PCIe lanes for the chipset, which similar to Intel acts like a PCIe switch for more IO - although AMD only offers a few lanes here compared to Intel's 24 PCIe 3.0 x4 lanes.

From the chipset there are eight PCIe lanes, which AMD here has suggested reserving two for gigabit Ethernet, four for a PCIe x4 slot, one for a PCIe x1 slot, and one for a WiFi module. Alongside the PCIe lanes are eight SATA ports, six USB 2.0 ports, four USB 3.1 (5 Gbps) Type-A ports, one USB 3.1 (5 Gbps) Type-C port and two USB 3.1 (10 Gbps) ports with appropriate redrivers. In a change to what we normally see on Intel motherboards, AMD lists SPI and Audio features as coming from the CPU rather than the chipset.

AMD Suggests Variants
As part of the launch, AMD is wanting to promote the different ways in which an X399 motherboard could be arranged depending on the workload requirements.

Each of these potential configurations would run into several thousand dollars a piece for a full build - the M.2 drives were suggested at 512GB minimum, with the SATA drives in RAID configurations as well. AMD is making no joke; they want to be the premium platform.

New Controllers for X399
With each of these motherboard overviews, we like to cover a few of the newer controllers to hit the market that motherboard manufacturers might take advantage of. Given the recent X299 launch on the Intel side, not much has changed since. The big parts to enter the consumer space are Aquantia's multi-gigabit ethernet controllers.

Aquantia launched two ethernet controllers: the AQC107 which supports 100MB, 1G, 2.5G, 5G and 10G, and the AQC108 which supports 100MB, 1G, 2.5G and 5G, all through standard RJ45 connectors. Motherboard manufacturers can use these controllers directly on the boards, or bundle them in add-in cards like the AQN107 and AQN108 above. We've seen cards from ASUS and GIGABYTE already hit the market based on these controllers, while ASRock and MSI prefer to integrate them direct.

The second controller worth mentioning is the updated ASMedia ASM3142 controller, providing USB 3.1 Gen 2 (10 Gbps) functionality to motherboards that need more USB 3.1g2. Originally we saw the ASM1142 single lane controller, while the ASM2142 integrated a dual lane design such that all the ports had peak bandwidth when multiple ports are in use. ASM3142 is a low powered version of the ASM2142, but still supporting all the ports at peak bandwidth.

Third is the updated Realtek ALC1150 audio codec, known as the ALC1220. The ALC1220 improves peak SNR while adding new functionality. ASUS run a customized version known as the ALC1220A, which removes some of the excess function for a stripped down version unique to them but configured in a partnership with Realtek.

The Motherboards
The seven motherboards we have to discuss are:

The ASRock X399 Taichi
The ASRock X399 Professional Gaming
The ASUS X399 ROG Zenith Extreme
The ASUS Prime X399-A
The ASUS Strix X399-E Gaming
The GIGABYTE X399-Gaming 7
The MSI X399 Gaming Pro Carbon

Covering the X399 Spectrum
We've seen details on seven boards which, on the surface, can seem similar as far as feature sets go. Owners that just want a Threadripper system to get up and go will get the necessities out of any motherboard listed here. The differences lay in the details: in networking, PCIe layout, RGB, overclocking features, enthusiast level gaming or compute, the want/need for 10 Gigabit ethernet, or storage options like fast M.2 and U.2. Even the type and number of USB ports can all factor into a buying decision.

It is also hard to forget about looks. All motherboards have some form of RGB LEDs scattered around their black PCBs. Some incorporate less than others, with the cheaper boards from ASRock and ASUS only applying them on the chipset heatsink, while others like the Gigabyte X399 AORUS Gaming 7 are loaded to the brim. Only one board, the ASUS X399 Prime, doesn't have wireless or Bluetooth capability. The MSI board sets itself apart from the rest by including bundled sets of vanity plates for the chipset heatsink, I/O and Audio covers, giving owners a bit more flexibility over the boards base appearance. That said, beauty is in the eye of the beholder. It is up to the buyer to figure out which looks best in their system vision. Assuming you care about aesthetics, that is.

PCIe differences come down to the number of slots for the GPUs, and if a user requires tri-slot spacing. For users looking at PCIe co-processors other than GPUs, then there isn't an ideal 7-slot motherboard on the market so far. But plug in a request, see if a manufacturer bites.
With Threadripper CPUs TDP set at 180W, an efficient VRM solution will be needed for cool operation, as well as for overclocking: many of the boards are using International Rectifier parts (MSI, GIGABYTE, ASRock) with the phase count ranging from eight to thirteen depending on the board. Keeping the VRMs cool is also a point to be considered, and the boards do this different. ASUS has two heatsinks connected via a heat-pipe, and include a fan for active cooling. The ASRock Professional Gaming, Taichi, and Gigabyte Aorus Gaming 7 also use two heat sinks but are passive (no fans). The MSI has a single VRM heatsink, but has doubled the standard height.

The table below shows several of the specifications differences between the released motherboards. The bold values are the top specifications for the category.

The table is clearly not the final discussion - each user has different requirements out of their system and motherboard. Some enthusiasts may want to overclock their boards more than others, or have a need for 10 gigabit Ethernet capabilities or the latest wireless networking options, or require an ATX form factor in their current chassis. Perhaps all of those, or none. We're likely to see further products hit the shelves in the coming months and quarters, especially as AMD has confirmed that this socket will stay for at least one more generation of products.

Pricing on the boards range from $340 (ASRock Taichi), up to $550 for the ASUS' flagship Zenith Extreme so there is quite a range already. All boards detailed here are available, with ASUS saying the Strix arriving later.

 

[-Fran-]

It's about time 10Gbps becomes a thing for LAN! When can we start expecting them to be integrated into the MoBo as part of the regular lines? haha

Cheers!

 

[goldstone77]

Yuka, I would expect them to find a form of wifi capable of faster speeds before that happens, or as that happens!

 

[8350rocks]

Yuka, I would expect them to find a form of wifi capable of faster speeds before that happens, or as that happens!

Intel abandoned WiGig because signal strength is too dependent upon outside variables. Wireless AC is pretty strong as it is now anyway...

However, I am very surprised that 10 Gbps has not become more common.

 

[goldstone77]

Yuka, I would expect them to find a form of wifi capable of faster speeds before that happens, or as that happens!

Intel abandoned WiGig because signal strength is too dependent upon outside variables. Wireless AC is pretty strong as it is now anyway...

However, I am very surprised that 10 Gbps has not become more common.

It depending on LOS it couldn't go through walls.... Waste of time and money if it has no penetration power. Wireless is going to be the key. Someone just has to figure out how to do it right. My cellphone works great!

 

[-Fran-]

Well... If you want signal strength, you can use Mhz bands with packaged signaling. High latency, very good bandwidth for good modulated frequencies. It's the same trick DSL uses, but applied to airwaves. That is why regular radio signals use them, but you do need a lot of space for them. waves in the Mhz range actually are 50cm+ in length, or something like that. So capturing them is kind of tricky for small things... Hence why they might not be the preferred range.

If you go into high Ghz, you run into 2 problems: signal power and noise. The 1Ghz to 10Ghz range is used a lot in different things, where 2.4Ghz and 5.2Ghz seem to be the most common. Plus, higher frequency demands a faster conversion, so the receivers would be more expensive I would imagine, plus processing the signal and all of that... You have the problem of materials affecting signals even more so, the noise gets more of a PITA to process at the receiver ends, plus packaging them and sending it, it's costly. It is directly proportional to the amount of data you want to transfer, since you still need to transform the data into chunks, compress them, modulate them, produce the transport signal and send them!

Ugh, I need to re-read this... It's been a while since I've toyed with radio signaling.

TL;DR: Wifi sucks! XD

Cheers!

 

[goldstone77]

Well... If you want signal strength, you can use Mhz bands with packaged signaling. High latency, very good bandwidth for good modulated frequencies. It's the same trick DSL uses, but applied to airwaves. That is why regular radio signals use them, but you do need a lot of space for them. waves in the Mhz range actually are 50cm+ in length, or something like that. So capturing them is kind of tricky for small things... Hence why they might not be the preferred range.

If you go into high Ghz, you run into 2 problems: signal power and noise. The 1Ghz to 10Ghz range is used a lot in different things, where 2.4Ghz and 5.2Ghz seem to be the most common. Plus, higher frequency demands a faster conversion, so the receivers would be more expensive I would imagine, plus processing the signal and all of that... You have the problem of materials affecting signals even more so, the noise gets more of a PITA to process at the receiver ends, plus packaging them and sending it, it's costly. It is directly proportional to the amount of data you want to transfer, since you still need to transform the data into chunks, compress them, modulate them, produce the transport signal and send them!

Ugh, I need to re-read this... It's been a while since I've toyed with radio signaling.

TL;DR: Wifi sucks! XD

Cheers!

What is the transfer rate wifi now? They are just using that multi-beaming to increase throughput.
ASUS AC5300 Wireless Tri-Band (Dual 5GHz + Single 2.4GHz) Gigabit Router [RT-AC5300] 802.11ac 5334 Mbps Wi-Fi Speed, MU-MIMO, 4 x 4 Antenna Design, 4x Gigabit LAN Ports, WTFast Route Optimization
Tri-band (dual 5 GHz, single 2.4 GHz) with the latest 802.11ac 4x4 technology for maximum throughput (5334 Mbps) and coverage (up to 5,000 sq. ft.)
MU-MIMO technology enables multiple compatible clients to connect at each client’s respective maximum speed
Built-in access to WTFast Gamers Private Network (GPN) of route-optimized servers ensures low, stable ping times for gaming
AiProtection Powered by Trend Micro provides multi-stage protection from vulnerability detection to protecting sensitive data
ASUS Smart Connect delivers consistent bandwidth by dynamically switching devices between 2.4 and 5 GHz bands based on speed, load and signal strength ; NOTE : Refer the user manual .
ASUS Ranked “Highest Customer Satisfaction with Wireless Routers in the U.S.”– J.D. Power

 

[-Fran-]

All those designs just increase bandwidth, but don't improve latency AFAIK. That is where wired is still king.

Cheers!

 

[8350rocks]

All those designs just increase bandwidth, but don't improve latency AFAIK. That is where wired is still king.

Cheers!

Yep, wired is instant, wifi induces lag.

 

[goldstone77]

All those designs just increase bandwidth, but don't improve latency AFAIK. That is where wired is still king.

Cheers!

Well, we are going to have to wait till they can master quantum entanglement to have lag free wireless!

 

[juanrga]

Confirmed that AMD lied us again, there are four dies on ThreadRipper, not two.

https://www.eteknix.com/threadripper-not-contain-dummy-dies/

This proves that ThreadRipper are failed EPYC and gives extra weight to my theory about the true origin of ThreadRipper from the fiasco of the SP4 platform.

 

[-Fran-]

Confirmed that AMD lied us again, there are four dies on ThreadRipper, not two.

https://www.eteknix.com/threadripper-not-contain-dummy-dies/

This proves that ThreadRipper are failed EPYC and gives extra weight to my theory about the true origin of ThreadRipper from the fiasco of the SP4 platform.

I see it just as extra harvesting. That is the whole point of their design. Since you don't know the timing of things, you will never be able to accurately determine how things went.

 

[goldstone77]

Confirmed that AMD lied us again, there are four dies on ThreadRipper, not two.

https://www.eteknix.com/threadripper-not-contain-dummy-dies/

This proves that ThreadRipper are failed EPYC and gives extra weight to my theory about the true origin of ThreadRipper from the fiasco of the SP4 platform.

Like I've already said, next year we get to see 32 core ThreadRipper at 7nm!

 

[juanrga]

Confirmed that AMD lied us again, there are four dies on ThreadRipper, not two.

https://www.eteknix.com/threadripper-not-contain-dummy-dies/

This proves that ThreadRipper are failed EPYC and gives extra weight to my theory about the true origin of ThreadRipper from the fiasco of the SP4 platform.

Like I've already said, next year we get to see 32 core ThreadRipper at 7nm!

There is no Zen on 7nm neither next year nor 2019. Therefore no...

 

[goldstone77]

Confirmed that AMD lied us again, there are four dies on ThreadRipper, not two.

https://www.eteknix.com/threadripper-not-contain-dummy-dies/

This proves that ThreadRipper are failed EPYC and gives extra weight to my theory about the true origin of ThreadRipper from the fiasco of the SP4 platform.

Like I've already said, next year we get to see 32 core ThreadRipper at 7nm!

There is no Zen on 7nm neither next year nor 2019. Therefore no...

That remains to be seen!