EPYC almost runs on X399

[Jito463]

1 hour ago, skankhunt said:

Technically they all use the same dies - just binned from EPYC -> Threadripper -> Ryzen etc. 

 

And the 32 core EPYCs use 4 CCX's. I.e. 4x two 4 core dies. 

One further correction, each die is comprised of two CCXs, therefore each Epyc chip has 8 CCXs (4 dies x 2 CCX/die = 8 CCX total).  

[Nicnac]

lol this is like "it works on X399!!" Except everything else doesn't

[Skanky Sylveon]

I have one word for this article.

 

Clickbait.

[BluJay614]

16 hours ago, Cinnabar Sonar said:

I have one word for this article.

 

Clickbait.

Okay, now explain how it is clickbait. Or do we have a confession of trolling?

[Jito463]

4 hours ago, BluJay614 said:

Okay, now explain how it is clickbait.

Because, despite the sensationalist nature of the title, Epyc really doesn't run on x399.  It "almost runs" in the same way that I'm "almost a millionaire".  'Almost' only counts in horseshoes and hand grenades.

[BluJay614]

7 minutes ago, Jito463 said:

Because, despite the sensationalist nature of the title, Epyc really doesn't run on x399.  It "almost runs" in the same way that I'm "almost a millionaire".  'Almost' only counts in horseshoes and hand grenades.

If you don't like it, don't click on it. The only one who made you click on it or read it was you.

[Jito463]

1 minute ago, BluJay614 said:

If you don't like it, don't click on it. The only one who made you click on it or read it was you.

Huh?  My reading the article or not doesn't alter whether it's a "clickbait" article.  I'm not necessarily blaming you per se, as it's primarily the fault of WCCFTech (though you did repost it).  Regardless of where it originated from, it still falls under the classification of clickbait.

 

Besides, I'm not the one who made the claim, I simply explained it.  Having said that, I do concur with that assessment.

[BluJay614]

1 minute ago, Jito463 said:

Huh?  My reading the article or not doesn't alter whether it's a "clickbait" article.  I'm not necessarily blaming you per se, as it's primarily the fault of WCCFTech (though you did repost it).  Regardless of where it originated from, it still falls under the classification of clickbait.

 

Besides, I'm not the one who made the claim, I simply explained it.  Having said that, I do concur with that assessment.

And I'm still going to disagree. Clickbait is when it's worded in a way to entice you to get you to click on a link or story, and the subject matter had nothing to do with it.

[cj09beira]

3 hours ago, Jito463 said:

Huh?  My reading the article or not doesn't alter whether it's a "clickbait" article.  I'm not necessarily blaming you per se, as it's primarily the fault of WCCFTech (though you did repost it).  Regardless of where it originated from, it still falls under the classification of clickbait.

 

Besides, I'm not the one who made the claim, I simply explained it.  Having said that, I do concur with that assessment.

i dont think its click bait at all considering that according to the experiments of derbaur a simple bios mod/update would allow the cpus to work on threadripper boards

[Jito463]

1 hour ago, cj09beira said:

i dont think its click bait at all considering that according to the experiments of derbaur a simple bios mod/update would allow the cpus to work on threadripper boards

That's just speculation, though.  None of his experiments prove that, they just show that it can power up the machine.  The system still didn't POST.  We have no idea if there's any specific instructions on the Epyc chips that aren't supported by X399 boards.

 

I refer to my previous post on the matter:

On 1/29/2018 at 2:08 PM, Jito463 said:

Though I still question whether a BIOS update actually would be enough.  Look at the work that went into getting an 8th Gen i3 running on a Z170 board, and then not everything functioned correctly.  This could be a similar situation.  Even if it's possible to get it to boot, not everything may work as expected.

I'm not saying it's impossible, I'm just saying that - based on the information presented - there's still far too much we don't know to be making any such assumptions.

[cj09beira]

5 minutes ago, Jito463 said:

That's just speculation, though.  None of his experiments prove that, they just show that it can power up the machine.  The system still didn't POST.  We have no idea if there's any specific instructions on the Epyc chips that aren't supported by X399 boards.

 

I refer to my previous post on the matter:

I'm not saying it's impossible, I'm just saying that - based on the information presented - there's still far too much we don't know to be making any such assumptions.

its the same silicon, same pinout, just with a Is epyc byte not posting with non registered eec ram is to be expected,

next thing to try is the correct ram , 

not posting is probably related to that (as the boot stops at the "memory light")

[Jito463]

Just now, cj09beira said:

its the same silicon, same pinout, just with a Is epyc byte

Another assumption.  I agree that's plausible, but we don't know that for certain.  And the reason I consider this article clickbait, is because it didn't "almost run".  It powered on the board, but nothing more.  That's not even close to running.

 

If the title had been something along the lines of "Experiments run with Epyc processor in X399 motherboard", I wouldn't have called it CB.  The title, however, suggests they almost got it running.  Powering on =/= running.

[cj09beira]

2 minutes ago, Jito463 said:

Another assumption.  I agree that's plausible, but we don't know that for certain.  And the reason I consider this article clickbait, is because it didn't "almost run".  It powered on the board, but nothing more.  That's not even close to running.

 

If the title had been something along the lines of "Experiments run with Epyc processor in X399 motherboard", I wouldn't have called it CB.  The title, however, suggests they almost got it running.  Powering on =/= running.

if the pinout was different there would have been flames and they would put some notches on the socket to prevent that, plus the cpu did do a good part of the booting process but got stuck on memory training which makes sense considering that epyc only supports registered ecc ram and he was using high speed non ecc ram

[Jito463]

23 minutes ago, cj09beira said:

if the pinout was different there would have been flames and they would put some notches on the socket to prevent that,

There doesn't have to be a different pinout for it to be incompatible.  There could be pins used by Epyc that are just dead pins on TR.  Yet again, you're speculating.

23 minutes ago, cj09beira said:

plus the cpu did do a good part of the booting process but got stuck on memory training

I can't watch the video right now, but regardless that doesn't mean it would progress any further just because it stopped there.

24 minutes ago, cj09beira said:

which makes sense considering that epyc only supports registered ecc ram and he was using high speed non ecc ram

So why didn't he try with ECC RAM?  Seems a simple fix if that's supposedly all it would take.  I have a hard time believing a high profile guy like "Der8aurer" couldn't get ahold of some ECC memory.  Heck, I could get ahold of some.

 

I don't have an issue with speculating whether Epyc would work in X399, I just take exception to making sensationalist claims like "it almost runs" without sufficient evidence.

[cj09beira]

7 minutes ago, Jito463 said:

There doesn't have to be a different pinout for it to be incompatible.  There could be pins used by Epyc that are just dead pins on TR.  Yet again, you're speculating.

I can't watch the video right now, but regardless that doesn't mean it would progress any further just because it stopped there.

So why didn't he try with ECC RAM?  Seems a simple fix if that's supposedly all it would take.  I have a hard time believing a high profile guy like "Der8aurer" couldn't get ahold of some ECC memory.  Heck, I could get ahold of some.

 

I don't have an issue with speculating whether Epyc would work in X399, I just take exception to making sensationalist claims like "it almost runs" without sufficient evidence.

cause we forgot about that detail, we in the comments reminded him afterwards 

[TopHatProductions115]

8 minutes ago, Jito463 said:

There doesn't have to be a different pinout for it to be incompatible.  There could be pins used by Epyc that are just dead pins on TR.  Yet again, you're speculating.

I can't watch the video right now, but regardless that doesn't mean it would progress any further just because it stopped there.

So why didn't he try with ECC RAM?  Seems a simple fix if that's supposedly all it would take.  I have a hard time believing a high profile guy like "Der8aurer" couldn't get ahold of some ECC memory.  Heck, I could get ahold of some.

 

I don't have an issue with speculating whether Epyc would work in X399, I just take exception to making sensationalist claims like "it almost runs" without sufficient evidence.

So, it's not clickbaiting. It's sensationalism of an approximate non-truth?

[Jito463]

18 minutes ago, TopHatProductions115 said:

So, it's not clickbaiting. It's sensationalism of an approximate non-truth?

[Crunchy Dragon]

Similar to the Athlon MP and XP processors, I'm guessing? Very interesting, I would also like to see this happen. The last exciting bit I saw in the CPU world was an i3-8350K on an MSI Z270 board.

[Drak3]

I almost slept with a model.

 

Funnily enough, my statement is closer to the truth.

[ARikozuM]

Just now, Drak3 said:

I almost slept with a model.

I'm guessing it was a Compaq 4540. You like them loud, don't you?

 

On topic: I'm guessing no one is giving AMD crap for this?

[Jito463]

45 minutes ago, ARikozuM said:

On topic: I'm guessing no one is giving AMD crap for this?

Why?  I'm not aware of anyone caring that the Xeons don't run on X299.  And if some Xeons do work, then I doubt there's very many which do.  HEDT isn't the server market, and vice-versa.

[mr moose]

54 minutes ago, ARikozuM said:

I'm guessing it was a Compaq 4540. You like them loud, don't you?

 

On topic: I'm guessing no one is giving AMD crap for this?

Not really, Although to be honest I am a bit surprised we haven't had one or to people trying to have a go. It's the sort of topic that's ripe for the misinformation that usually leads to long posts regarding sockets and how companies should give me what I want because they make stuff.

[Skanky Sylveon]

19 hours ago, BluJay614 said:

do we have a confession of trolling?

Trust me, you would know if I were trolling you.

19 hours ago, BluJay614 said:

explain how it is clickbait.

The definition of clickbait.

"something (such as a headline) designed to make readers want to click on a hyperlink especially when the link leads to content of dubious value or interest"

 

So, when an article has in the title.

"Hurr durr AMD EPYC Successfully Gets Working on a Ryzen Threadripper Motherboard After Small Modifications"

 

When it's not working (I mean for fucks sake, it's not even POSTing).  Then the article is being misleading.

It's making it sound more exciting then it actually is, and thus, may lead to more clicks then if it's title was more truthful.

[cj09beira]

2 hours ago, Cinnabar Sonar said:

 

Why are we using wccftech anyways the video from derbaur has all the info without the bad journalism ?

[Skanky Sylveon]

1 minute ago, cj09beira said:

Why are we using wccftech anyways the video from derbaur has all the info without the bad journalism ?

Agreed 100%, der8auer's video gives far more information.  For anyone who hasn't watched it yet, I highly recommend it.