• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD VEGA confirmed for 2017 H1

Status
Not open for further replies.
Gregster said:
According to their update on Guru, they also feel it is true, so not sure why you are getting so irate over someone posting a possible leak, true or not!
Click to expand...

So let's sum up here. Guru now thinks that's true.

That means this "Vega" card is 6% faster than the stock Fury X according to their original post.
This is when clocked 1200mhz vs 1050mhz.

So that means Vega has regressed in IPC.

We know the Instinct MI25 which is Vega has 12.5TFlops. That means it has a clock speed of 1500Mhz, and it's a passive air cooled accelerator just like Tesla.

We know HBM2 is rated for up to 1000Mhz, that's for 1.6Gbps and 2.0Gbps modules, which are the ones AMD will be using.

NVIDIA are using 1.4Gbps modules, which are clocked at 715Mhz.

Yet "Vega" in that Time spy link shows memory at 700Mhz, which is slower than the slowest available HBM2 modules.

The Vega die is almost the same since as Fiji, while the former is on 14nm compared to 28nm.
So Vega is a much larger chip.


So apparently, gaming Vega, which is going to be Water Cooled as a premium part, is going to have under 10Tflops, and be significantly slower clocked compared to an air cooled server part.
While having somehow lower clocked HBM2 when using the higher bandwidth and higher clocked modules.

All this, and then despite Vega being a larger GPU, and all the architectural improvements means Vega has severely regressed in IPC and overall performance compared to two year old Fiji.

That doesn't add up for me.
 
Last edited:
N19h7m4r3 said:
So let's sum up here. Guru now thinks that's true.

That means this "Vega" card is 6% faster than the stock Fury X according to their original post.
This is when clocked 1200mhz vs 1050mhz.

So that means Vega has regressed in IPC.

We know the Instinct MI25 which is Vega has 12.5TFlops. That means it has a clock speed of 1500Mhz, and it's a passive air cooled accelerator just like Tesla.

We know HBM2 is rated for up to 1000Mhz, that's for 1.6Gbps and 2.0Gbps modules, which are the ones AMD will be using.

NVIDIA are using 1.4Gbps modules, which are clocked at 715Mhz.

Yet "Vega" in that Time spy link shows memory at 700Mhz, which is slower than the slowest available HBM2 modules.

The Vega die is almost the same since as Fiji, while the former is on 14nm compared to 28nm.
So Vega is a must larger chip.


So apparently, gaming Vega, which is going to be Water Cooled as a premium part, is going to have under 10Tflops, and be significantly slower clocked compared to an air cooled server part.
While having somehow lower clocked HBM2 when using the higher bandwidth and higher clocked modules.

All this, and then despite Vega being a larger GPU, and all the architectural improvements means Vega has severely regressed in IPC and overall performance compared to two year old Fiji.

That doesn't add up for me.
Click to expand...

It doesn't for anyone applying some logic to the argument, however clickbaiting is big business today.
 
My only concern is that we're getting no Vega leaks that blow our doors off, only on the lower end of the performance expected. I mean, not really much reason to sandbag nVidia now that the Ti is out is their??? Hopefully we get some on the higher end soon, or AMD are just tempering our expectations somewhat....
 
JediFragger said:
My only concern is that we're getting no Vega leaks that blow our doors off, only on the lower end of the performance expected. I mean, not really much reason to sandbag nVidia now that the Ti is out is their??? Hopefully we get some on the higher end soon, or AMD are just tempering our expectations somewhat....
Click to expand...

Yeah could be a good thing or a bad thing when it comes to having no leaks...

gut feeling.. we will start getting interesting Darth Vega leaks on May the 4th though :D:p
 
JediFragger said:
My only concern is that we're getting no Vega leaks that blow our doors off, only on the lower end of the performance expected. I mean, not really much reason to sandbag nVidia now that the Ti is out is their??? Hopefully we get some on the higher end soon, or AMD are just tempering our expectations somewhat....
Click to expand...

Find me an AMD card that had proper information leak the last 5 years. None.
Always speculation.
And those guru3d, wcc****, tech<whatever>, still showing "new" articles with the top vega as the upcoming RX 490 card, weeks after AMD said that the name is RX Vega and no numbers.

So their credibility is pants these days.
 
Last edited:
Its possible that there is some Vega part (most likely pre-production engineering) that performs in the 1070 - 1080 space. So what, that tells you nothing a big fat zero because you have to piece so many assumptions together its pure speculation as to whether you are accurate or not. The important point is what the high end Vega parts for gaming are capable of and there is no way its going to perform like a 1070 that is a ridiculous assertion even for some one like me running an Nvidia gpu atm. This is all just ****** news trying to provoke a reaction from readers.
 
Panos said:
Find me an AMD card that had proper information leak the last 5 years. None.
Always speculation.
And those guru3d, wcc****, tech<whatever>, still showing "new" articles with the top vega as the upcoming RX 490 card, weeks after AMD said that the name is RX Vega and no numbers.

So their credibility is pants these days.
Click to expand...

No I understand, it's just that we're getting a little desperate now for *something* lol!

I'll glad make the switch back but I need it to be as fast as a Ti in DX12/Vulkan, I'm not expecting as much in DX11 as the 10-series are pretty much DX11 cards with a bit of DX12 functionality thrown in (whereas the Vega line will be pretty much fully DX12) :)
 
N19h7m4r3 said:
So let's sum up here. Guru now thinks that's true.

That means this "Vega" card is 6% faster than the stock Fury X according to their original post.
This is when clocked 1200mhz vs 1050mhz.

So that means Vega has regressed in IPC.

We know the Instinct MI25 which is Vega has 12.5TFlops. That means it has a clock speed of 1500Mhz, and it's a passive air cooled accelerator just like Tesla.

We know HBM2 is rated for up to 1000Mhz, that's for 1.6Gbps and 2.0Gbps modules, which are the ones AMD will be using.

NVIDIA are using 1.4Gbps modules, which are clocked at 715Mhz.

Yet "Vega" in that Time spy link shows memory at 700Mhz, which is slower than the slowest available HBM2 modules.

The Vega die is almost the same since as Fiji, while the former is on 14nm compared to 28nm.
So Vega is a much larger chip.


So apparently, gaming Vega, which is going to be Water Cooled as a premium part, is going to have under 10Tflops, and be significantly slower clocked compared to an air cooled server part.
While having somehow lower clocked HBM2 when using the higher bandwidth and higher clocked modules.

All this, and then despite Vega being a larger GPU, and all the architectural improvements means Vega has severely regressed in IPC and overall performance compared to two year old Fiji.

That doesn't add up for me.
Click to expand...

Might just be one of my favourite posts and responses on this forum. Hat off to you sir. Straight talk, logic and reasoning make for great contributions.

Always amazes me how long these new card threads go on for and as time goes on, the ratio of useful posts to noise decreases...not that I'm blaming...I know how being an enthusiast can make anyone less rational than usual...wish AMD would just put everyone out of their misery though...Computex seems a very long time away...
 
Joe_Easy said:
Might just be one of my favourite posts and responses on this forum. Hat off to you sir. Straight talk, logic and reasoning make for great contributions.

Always amazes me how long these new card threads go on for and as time goes on, the ratio of useful posts to noise decreases...not that I'm blaming...I know how being an enthusiast can make anyone less rational than usual...wish AMD would just put everyone out of their misery though...Computex seems a very long time away...
Click to expand...

AMD are well awake to, and tired of, nVidia sandbagging and trumping released products. They are an order of magnitude tighter about things than they previously were and who could blame them?

IMO due to this nVidia has released 2 titans and the 1080/ti all in the same generation. AMD are saying nada and nVidia are trying to fish with dynamite. Telling. We will see what happens this month I think.
 
A 1450Mhz Fury X already goes toe to toe with a 2ghz 1080 in firestrike extreme.
original article

And you have to consider that Fury X performance should be a baseline for Vega performance.

get


2.1GHz 1080 score link
 
Last edited:
Mauller said:
A 1450Mhz Fury X already goes toe to toe with a 2ghz 1080 in firestrike extreme.
original article

And you have to consider that Fury X performance should be a baseline for Vega performance.

get


2.1GHz 1080 score link
Click to expand...

And that is Fury X (an old 28nm GPU)... They have made many changes and improvements in Vega. If those bear fruit then we can easily expect Titan XP performance like predicted :D
 
Last edited:
JediFragger said:
Panos is getting irritated :D

Ever thought of being an AMD rep? It'd suit you ;) :p
Click to expand...
Not enough money :) I prefer to be software developer contractor.
(to afford the current pricing lol :p)


But yeah I am getting irritated, when we have a discussion going for pages, with good analysis, and someone comes in throws an article and goes off.
 
Last edited:
The situation with that Vega Timespy score, is that we don't know the state of the part being tested, if it is fully functional or not. Or even if it is an actual vega part.

The other side of things is to do with the drivers, is it running a for Vega DX12 driver? or is it using some fiji DX12 driver with emulation and debug enabled etc?
 
Mauller said:
The situation with that Vega Timespy score, is that we don't know the state of the part being tested, if it is fully functional or not. Or even if it is an actual vega part.

The other side of things is to do with the drivers, is it running a for Vega DX12 driver? or is it using some fiji DX12 driver with emulation and debug enabled etc?
Click to expand...

What ever it is, is not representative at all of what we know of Vega perf.
Is damn a lot slower than the passive cooled Vega MI25 card, and slower than a barely overclocked FuryX.

It could a engineering sample from Volta for all we know. Is just numbers without any substance.

And common sense. Gibbo couldn't have come in here, in this very thread, writing that he is in Rome, and Amd is in Rome, and what he sees is very impressive. If it had that performance....

Imho NV it could be even marketing department is trying to throw mud, and put people off from waiting few weeks for the GPU.
 
Last edited:
Panos said:
NO is not true.Read the last couple of pages please.
Click to expand...

hence the "sorry if this has been posted already by another website" - simple answer is "yes been posted a few pages back" - If you've stated no- and its turns out to be True, BUT was on an early Engineering Sample ( just like Ryzen engineering samples were god aweful and why we had the delay of X370 motherboards) oR non optimized drivers you make your self look like a Jackass :D

I know this first hand as I've given dates on here that have been written down in my NDA which turns out to be an older document

So may apologies to thread viewers of a duplicate article post :)
 
Status
Not open for further replies.
Back
Top Bottom