AMD making a fool of Threadripper customers - AGAIN?

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

NablaSquaredG

Layer 1 Magician
Aug 17, 2020
1,320
800
113
A bit of backstory first:
X399 was discontinued after just 1.5 generations (Zen1 Threadripper 1000 and Zen1+ Threadripper 2000).
There was an outrage in the PC community once it came to light that Zen2 Threadripper 3000 will NOT be compatible with existing X399 motherboard.

AMD tried to justify by saying that X399 wouldn't have allowed Threadripper 3000 to unleash its full potential due to the chipset being connected with PCIe3.0 and VRM (TR 1000 had a TDP of 180W and 2000 up to 250W, whereas TR 3000 has 280W) stuff (which is true) and that launching a new platform would result in a long-term stable platform

You can read some of the things here on reddit https://www.reddit.com/r/Amd/comments/dsy4kw (Post from the official AMD Account)

Money quote:
The socket change also sets us up nicely for future development and scalability of the Threadripper platform, both on a near- and long-term basis.
Now:
According to multiple sources, there are rumors that Threadripper 5000, aka (ex) Chagall, is cancelled and that AMD will ONLY Release Threadripper PRO 5000.
Threadripper PRO is not compatible with the standard Threadripper sTRX4 Platform.

If that is true (which I'm afraid seems quite likely given AMD's decisions in the last 1.5 years), AMD would not only have done the same thing again, aka (needlessly) producing a ton of extremely expensive (Threadripper boards ARE expensive) e-waste, but rather make an even bigger fool of all the loyal customers who believed in their (probably intentionally a bit vague) promise and cancel the sTRX4 generation even earlier, after exactly one generation.

It is sad to see how AMD develops, but this seems to line up nicely with their recent decisions (Vendor Locking EPYC and the last-minute decision to make Threadripper PRO non-overclockable)


What do you think about this?
 
  • Like
Reactions: ari2asem

alex_stief

Well-Known Member
May 31, 2016
884
312
63
38
The excuses didn't add up from the start. Every X399 board on the market could have easily handled higher TDP CPUs. They were all designed with overclocking in mind on the original CPUs, i.e. oversized power delivery.
Maybe people can slowly start seeing AMD for what it is, and always has been. A company like every other, obligated to create value for shareholders and partners. I do not see this as a change in company culture, but the end-game of a rather successful marketing campaign.
 
Last edited:

Wasmachineman_NL

Wittgenstein the Supercomputer FTW!
Aug 7, 2019
1,872
617
113
And in this episode of "CORPORATIONS ARE NOT YOUR FRIEND", people find out AMD and board partners care about money and discontinue Threadripper after two or even only one generation. Not the first time they pulled shit like this though *cough* discontinuing Zen 3 support for 300 series chipsets *cough*

Massive dick move by them of course, I agree, and the reason I didn't go with TR when building a new computer early this year.
 
  • Like
Reactions: gb00s and fohdeesha

ectoplasmosis

Active Member
Jul 28, 2021
117
53
28
Vanilla Threadripper was doomed from the moment the decision was made to release TR-Pro as a consumer part instead of limiting it to OEM pre-built workstations as it was first intended. The two platforms are needlessly fragmented, making little sense for them to coexist.

AMD also has a more formidable opponent in the form of the imminent Sapphire Rapids-based workstation platform, likely needing the 8-channel memory to compete.

It’s a logical move to can Threadripper.
 

ca3y6

New Member
Apr 3, 2021
24
3
3
Stupid question: do lots of people upgrade their desktop CPU without upgrading their motherboard? CPUs are so powerfull now that for desktop workloads, including gaming, they are rarely at capacity, unless you do stuff like compressing videos. Feels like it must be a tiny market.
 

NablaSquaredG

Layer 1 Magician
Aug 17, 2020
1,320
800
113
Stupid question: do lots of people upgrade their desktop CPU without upgrading their motherboard? CPUs are so powerfull now that for desktop workloads, including gaming, they are rarely at capacity, unless you do stuff like compressing videos. Feels like it must be a tiny market.
They definitely do if they can! Look at Desktop Ryzen and the amount of outrage there was when certain older chipsets didn't support newer CPUs.

Threadripper Boards are considerable more expensive than Ryzen boards (500 - 1000€), so it would make even more sense to have upgradability on those boards.
Not to mention the environmental impact of producing tons of expensive e-waste that could have worked one generation longer.
Also, see Intel: People got CPUs running on unsupported older chipsets with modified BIOS versions, so it is obvious that it would be possible to have upgradable boards.
 

i386

Well-Known Member
Mar 18, 2016
4,221
1,540
113
34
Germany
What do you think about this?
AMD had their plans/roadmaps.
Than came covid-19 and with it lock-downs that caused a short supply of electronics (required to produce more electronics :D).
AMD and everybody else has to adapt to this situation and make the best out of it. And the best is probably not to waste (manufacturing) resources for a small market ._.
 

ServerSemi

Active Member
Jan 12, 2017
130
34
28
I got my 3960x when it came out and I'm low key mad that they might not offer me an upgrade path with my Asus trx40 motherboard. This is what happens when there is no competition but I will vote with my wallet and if intel releases a Threadripper competitor I will purchase it to stick it to AMD.
 
  • Like
Reactions: NablaSquaredG

111alan

Active Member
Mar 11, 2019
290
107
43
Haerbing Institution of Technology
For the original enterprise SP3 socket, it theoretically supports all 3 gens of EPYC(But most vendors only let each board support two gens, due to BIOS flash capacity issue)

For the desktop/WS version, let's see, x399, trx40, WRX80 are all especially made not to support the next batch of CPUs(Zen1, Zen2-X, Zen2-WX).

It's just that what customer decides what product being made. Large enterprises have their own complexed and reliable test methologies so they know exactly what AMD's unbalanced designs are good and bad at, thus made them rational and hard to bamboozle. But on the outside AMD fans are famously notorious, fabricating test data and bending truths with their explanation to their needs, while swarm attacking everyone who have second thoughts. There are also a lot of evidences indicating AMD paying independent reviewers and commenters for its reputation campaign. If you've read "the crowd", you should already know why they did all of this, and why some people just buy it.

Just miss the time when flagship MSDT/HEDT CPUs only cost 400/1000$.
 

Wasmachineman_NL

Wittgenstein the Supercomputer FTW!
Aug 7, 2019
1,872
617
113
But on the outside AMD fans are famously notorious, fabricating test data and bending truths with their explanation to their needs, while swarm attacking everyone who have second thoughts. There are also a lot of evidences indicating AMD paying independent reviewers and commenters for its reputation campaign.
 

edge

Active Member
Apr 22, 2013
203
71
28
My corporate experience has taught me to buy what I need now and then replace with a current system when my old system isn't meeting my needs.

The idea of just upgrading one component to get better performance misses that computers are systems (Von Neumann architectures) and upgrading just one component leads to that component being bottlenecked by the unupgraded components. Eliminating one bottleneck just reveals another.

The advantage of rip and replace is you get to take advantage of the price performance curve - any level of performance decreases in price over time. For tech it has been an order of magnitude of performance increase at the same price every 5 years. This means buying tech with the view of upgrading later means you are paying a premium for technology that will be outdated when you go to upgrade it.

It may just be the op is projecting.
 

111alan

Active Member
Mar 11, 2019
290
107
43
Haerbing Institution of Technology
Can't recall everything that happened in the two years, but here are the direct and indirect proof I have at hand now, mostly in my country:

Direct:

1. One of my long time friend turned to work on drivers for AMD platforms, end up worked under an "AMD temporary worker". He asked me to remove a post mentioning him in a topic that's considered “negative” to AMD on Chiphell.

I will not show his name though, he is not the center of AMD propaganda. But through him and some other people I know that there are a lot of these "negative" post on Chiphell and NGA got deleted or locked with no explanation, including mine.
1.jpg

2. Just go to Bilibili, open any video concerning CPU posted before the release of intel 12th gen. Just see the sheer amount of "AMD YES" in the barrages and comments, and how many people use overclocked AMD cpu against intel's stock or even downed clock. Sometimes you will see people secretly down/over clocking memory or bus frequency to make up results.

3. Speaking of reviews, I'm not a believer of north bridge design when it comes gaming, or sayings like "intel performance regress over generations". For this, once, after I saw the reviews, I borrowed cpus and motherboards to test out myself.

Then there comes the problem. My results met my expectation and almost the same as Techpowerup's(AMD Ryzen 7 5800X Review). But why there were an attack on them and forced them to apologize and explain the situation in an AMD favored way?

Here's one of my graphs, all locked at 4GHz.
2.png
The same type of organized attack happend to another reviewer I know, named "Code Yuri", who claimed 11th gen core is better than 10th gen overall in gaming with a lot of tests backing, claiming him cheater and intel-payed, as well as one of my friend named 神经君 who tried to explain why he did not use AMD platform to test SSDs.

Indirect:

1. Another interesting phenomenon, everyone stress only "League of Legends" or "CS:GO" test results(which are hundreds even thousands of FPS) and disregard any other games, to say AMD has better gaming performance. Everyone use only Cinebench and CPUZ to talk about IPC, both are not common benchmarks before(they even didn't make it onto HWBOT's record list until recent years) and don't actually represent realworld IPC due to the research. They all talk about the same thing and share the same opinion with the exact same type of comments like "AMD YES""Intel will bankrupt""intel can not create any new things anymore". Coincidence?

2. I know Der8auer and Stilt experienced the same type of problem before. Der8auer's explaining video can still be found on Youtube, and Stilt's disabled Anand forum account and his last posts are still visible too. There are just too many personal attacks and false claims for me to believe that it's not organized.

3. Before I know this, I was also personal attacked on Chiphell as well. They used fake benchmarks(an SSD test results that's far faster than official speed),"I don't understand this but what's your purpose" and just fabricated what I did to guess my intentions. They came in turns with an admin locking the posts immediately when those people began to lose the argument. They are very organized. Perhaps I'll try law-suiting some of them if I meet the same situation again.

My username is 镜音リン on Chiphell. You can search my posts, along with other things I mentioned in this post. Sorry for not being able to sink hours browsing old records and post every single evidence, but in the future, if it's still need I may consider doing that.
 

Wasmachineman_NL

Wittgenstein the Supercomputer FTW!
Aug 7, 2019
1,872
617
113
Can't recall everything that happened in the two years, but here are the direct and indirect proof I have at hand now, mostly in my country:

Direct:

1. One of my long time friend turned to work on drivers for AMD platforms, end up worked under an "AMD temporary worker". He asked me to remove a post mentioning him in a topic that's considered “negative” to AMD on Chiphell.

I will not show his name though, he is not the center of AMD propaganda. But through him and some other people I know that there are a lot of these "negative" post on Chiphell and NGA got deleted or locked with no explanation, including mine.
View attachment 20939

2. Just go to Bilibili, open any video concerning CPU posted before the release of intel 12th gen. Just see the sheer amount of "AMD YES" in the barrages and comments, and how many people use overclocked AMD cpu against intel's stock or even downed clock. Sometimes you will see people secretly down/over clocking memory or bus frequency to make up results.

3. Speaking of reviews, I'm not a believer of north bridge design when it comes gaming, or sayings like "intel performance regress over generations". For this, once, after I saw the reviews, I borrowed cpus and motherboards to test out myself.

Then there comes the problem. My results met my expectation and almost the same as Techpowerup's(AMD Ryzen 7 5800X Review). But why there were an attack on them and forced them to apologize and explain the situation in an AMD favored way?

Here's one of my graphs, all locked at 4GHz.
View attachment 20940
The same type of organized attack happend to another reviewer I know, named "Code Yuri", who claimed 11th gen core is better than 10th gen overall in gaming with a lot of tests backing, claiming him cheater and intel-payed, as well as one of my friend named 神经君 who tried to explain why he did not use AMD platform to test SSDs.

Indirect:

1. Another interesting phenomenon, everyone stress only "League of Legends" or "CS:GO" test results(which are hundreds even thousands of FPS) and disregard any other games, to say AMD has better gaming performance. Everyone use only Cinebench and CPUZ to talk about IPC, both are not common benchmarks before(they even didn't make it onto HWBOT's record list until recent years) and don't actually represent realworld IPC due to the research. They all talk about the same thing and share the same opinion with the exact same type of comments like "AMD YES""Intel will bankrupt""intel can not create any new things anymore". Coincidence?

2. I know Der8auer and Stilt experienced the same type of problem before. Der8auer's explaining video can still be found on Youtube, and Stilt's disabled Anand forum account and his last posts are still visible too. There are just too many personal attacks and false claims for me to believe that it's not organized.

3. Before I know this, I was also personal attacked on Chiphell as well. They used fake benchmarks(an SSD test results that's far faster than official speed),"I don't understand this but what's your purpose" and just fabricated what I did to guess my intentions. They came in turns with an admin locking the posts immediately when those people began to lose the argument. They are very organized. Perhaps I'll try law-suiting some of them if I meet the same situation again.

My username is 镜音リン on Chiphell. You can search my posts, along with other things I mentioned in this post. Sorry for not being able to sink hours browsing old records and post every single evidence, but in the future, if it's still need I may consider doing that.
Damn dude, what I posted was a joke and then he delivers very well detailed proofs.

Interesting writeup, thanks.
 
  • Like
Reactions: ecosse

msg7086

Active Member
May 2, 2017
423
148
43
36
Everyone use only Cinebench and CPUZ to talk about IPC, both are not common benchmarks before
Was that why Intel used MS Office for benchmarking?

The AMD YES comments appeared way before 12gen was released. It probably started when first gen Ryzen was released. Fun story, at that time I thought the product would be a huge success so I convinced one of my co-workers to buy its stock and he did. It was about $10 a share when he bought it.

I upgraded from i7-4770 to 3600. The old 4770 platform was problematic all the time. It had 2 out of 4 memory slots dead. I got a new board shipped to me and it behaved the same. It's either a bad CPU or a bad motherboard batch (Asrock if you are interested). The 3600 platform is giving me the most value, and in a few years I'll be upgrading it to a 5900. It was a shame that i7-4770 had no upgrade path at all.

Both companies made crap products in the past. I pick Athlon64 when Intel made the hot pot P4. I went Intel core series when AMD got me the low efficiency construction machines. Now I'm going back Ryzen because what Intel was doing all these days?

It's interesting to see propaganda from BOTH sides, but at the end of the day the majority would buy those that maximize the value.

Now back to the topic.

It's sad to see AMD not releasing new TR platforms, presumably because Intel stopped competing at this market as well. As people always say, thanks to AMD we got better Intel products, and thanks to Intel we got better AMD products. When competition stops, this is what we get.

Not just AMD but also Intel. Still remember the days where E3-1230 can be had for a reasonable price?
 
  • Like
Reactions: leon_pro and gerby

WANg

Well-Known Member
Jun 10, 2018
1,302
967
113
46
New York, NY
A bit of backstory first:
X399 was discontinued after just 1.5 generations (Zen1 Threadripper 1000 and Zen1+ Threadripper 2000).
There was an outrage in the PC community once it came to light that Zen2 Threadripper 3000 will NOT be compatible with existing X399 motherboard.

AMD tried to justify by saying that X399 wouldn't have allowed Threadripper 3000 to unleash its full potential due to the chipset being connected with PCIe3.0 and VRM (TR 1000 had a TDP of 180W and 2000 up to 250W, whereas TR 3000 has 280W) stuff (which is true) and that launching a new platform would result in a long-term stable platform

You can read some of the things here on reddit https://www.reddit.com/r/Amd/comments/dsy4kw (Post from the official AMD Account)

Money quote:


Now:
According to multiple sources, there are rumors that Threadripper 5000, aka (ex) Chagall, is cancelled and that AMD will ONLY Release Threadripper PRO 5000.
Threadripper PRO is not compatible with the standard Threadripper sTRX4 Platform.

If that is true (which I'm afraid seems quite likely given AMD's decisions in the last 1.5 years), AMD would not only have done the same thing again, aka (needlessly) producing a ton of extremely expensive (Threadripper boards ARE expensive) e-waste, but rather make an even bigger fool of all the loyal customers who believed in their (probably intentionally a bit vague) promise and cancel the sTRX4 generation even earlier, after exactly one generation.

It is sad to see how AMD develops, but this seems to line up nicely with their recent decisions (Vendor Locking EPYC and the last-minute decision to make Threadripper PRO non-overclockable)


What do you think about this?
That Threadripper Chagall 5000 cancellation has been rumored since October, and it’s just that…rumors. Who knows what AMD is planning to do in 2022.

AMD didn't dick over their customers once - they did it at least three times before - the AM2 to AM2+ transition (theoretically compatible, but not that many vendors roll out firmware to do it), the AM2+ to AM3 (you can use an AM2+ on an AM3 board, but forward compatible? eeeeeeh), and the Socket 1207, which is some HEDT crap they pushed out for Quad FX (AMD K8 arch) in 2006 based on Socket F - not much people bought into it, but those who did, they got burnt for a dual dual socket setup that's not nearly as flexible as dual Opteron, and was instantly obsolete when the K10 Phenoms came out in 2007 - still not quite fast enough to match a C2Q though. So once again, not new, but also not a trait unique to AMD. As an AMD customer I stay away from their expensive stuff (anything HEDT depreciates quick anyways, they don't get the new features like their mobile lines, and I would rather just go whole hog with Eypc embedded for home use anyways), and as an AMD shareholder, I bought them at 6 back in the Phenom days, and it's now at 153 - seems great, but then the missus held nVidia stock since the FX days, and that kind of value comes from each company's abilities to dick over customers, year over year - enough to pay the bills but not enough to entirely drive them away - if you think AMD is bad, look at how nVidia does it. I still laugh at how nVidia forces end users to spend extra money to get GPU virtualization going, while Intel claims to have it for free since the Iris Pro days.

That AMD Eypc chip vendor lockout thing is a trusted computing thing - not every vendor will do it on the chips shipped with their machines. SuperMicro doesn't, and neither does HPE.

Was that why Intel used MS Office for benchmarking?

The AMD YES comments appeared way before 12gen was released. It probably started when first gen Ryzen was released. Fun story, at that time I thought the product would be a huge success so I convinced one of my co-workers to buy its stock and he did. It was about $10 a share when he bought it.

I upgraded from i7-4770 to 3600. The old 4770 platform was problematic all the time. It had 2 out of 4 memory slots dead. I got a new board shipped to me and it behaved the same. It's either a bad CPU or a bad motherboard batch (Asrock if you are interested). The 3600 platform is giving me the most value, and in a few years I'll be upgrading it to a 5900. It was a shame that i7-4770 had no upgrade path at all.

Both companies made crap products in the past. I pick Athlon64 when Intel made the hot pot P4. I went Intel core series when AMD got me the low efficiency construction machines. Now I'm going back Ryzen because what Intel was doing all these days?

It's interesting to see propaganda from BOTH sides, but at the end of the day the majority would buy those that maximize the value.

Now back to the topic.

It's sad to see AMD not releasing new TR platforms, presumably because Intel stopped competing at this market as well. As people always say, thanks to AMD we got better Intel products, and thanks to Intel we got better AMD products. When competition stops, this is what we get.

Not just AMD but also Intel. Still remember the days where E3-1230 can be had for a reasonable price?
MS Office, Adobe whatever suite, 7zip, handbrake, kernel compiles and who knows what else. I mean, you want your CPU to give you performance metrics that are similar to how people tend to use them, right? Otherwise you can spend all day hammering CrystalDiskMark, geekbench, CPU-Z and do bar graph measuring contests on your specific machine and spend hours upon hours arguing about the results on the internet. It's like watching people on car forums foaming at their mouths arguing over intake liners or suspension geometry or how Toyota engineers rob them of their precious seconds racing to their local 7-Elevens or how their Mustang is best Mustang (TM). Its pedantry at its worst.

Ah yes, the cheap and simple Xeon E3s back before Scalable is a synonym for "custom tailored to extract appropriate amounts of cash money from Intel customers", when you don't need an @Patrick and his Audubon guide to spotting Xeon processors? Seriously? Silver plumage, 6xxx series Genus, order data center, 6243QS with crispy fried onions, bacon toppings and AI tensor cores, and I have to consult a chart with 90+ members, multiple tiers and bins?

The earthmover series were crap, much like the Pentium 4. But then AMD and Intel has gone back and forth for decades. K6 II/III versus Pentium II/III, Athlon XP-m versus Intel Pentium-M, K7/K8 versus Pentium 4/Core, and the nonsense from both side dates back decades, like Intel suing AMD for making AM386/486, Intel paying ODMs to shut out AMD on laptops, or AMD getting Intel to adopt AMD64 (because the Itanic saga was hilarious, sad and a repeat of their iAXP432 disaster), or Intel shooting themselves in the foot with that RAMBUS saga. The list goes on and on and on and on. At the end of the day it's still about perceived value, and getting hardware that works best for you. AMD might not game as well as Intel on those big 16 core chips (although on the APU versus Intel integrated space a Ryzen 4000 based laptop will beat a Tiger Lake i5 with Xe G7 graphics), but not everyone waste cores idle for games that don't multithread well.
 
Last edited:
  • Like
Reactions: gerby

msg7086

Active Member
May 2, 2017
423
148
43
36
Intel shooting themselves in the foot with that RAMBUS saga
Haha now that you mentioned I almost forgot the rambus days. My dad got a pair of rambus with his Pentium 4 after he upgraded from a Pentium 133. It was a terrible build at the time.