-
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
- Thread starter Doom112
- Start date
- Status
I like the "DISPLAY CONECTORS None" bit heh.
Its "Alreet Guys"
Anyway, this guy has been accused of being an AMD fanboy, especially around here, i don't think so...
Once again its just good solid info, GTX 1080+ 10% is fine by me, it means the cut down Vega will probably be GTX 1070 + 10% and not hugely expensive.
Bring it
Yep. I'd actually say that would be a good result.
Yes, it's not a graphics card per say; and is used in HPC, essentially for acceleration and number crunching; Which is why they need lots of bandwidth as well. They also have lower clocks than the Titan at 1328Mhz, with a max boost of 1480Mhz
Stability and reliabilty over "gaming" performance
No fan either, as they're all in massive racks where sever fans blow air through them.
They don't even use standard drivers like a Quadro or GeForce card would.
Last edited:
That is what i said from the start that, AMD cannot even properly beat GTX 1080 with their upcoming GPU after 9- 10 months and they are aim for Volta in their marketing. That is the reason why AMD product fails to deliver because they never ever set the exception of their card honest and real.
Even AMD and AdoredTV said that Doom is best case scenario for VEGA performance and GTX 1080 is matching the performance on the same map.
Even AMD and AdoredTV said that Doom is best case scenario for VEGA performance and GTX 1080 is matching the performance on the same map.
Permabanned
- Joined
- 12 Sep 2013
- Posts
- 9,221
- Location
- Knowhere
Its "Alreet Guys"
Anyway, this guy has been accused of being an AMD fanboy, especially around here, i don't think so...
Once again its just good solid info, GTX 1080+ 10% is fine by me, it means the cut down Vega will probably be GTX 1070 + 10% and not hugely expensive.
Bring it
Performance wise that'll be spot on and it'll be down to the pricing, I'd like to buy big Vega but if it's stupidly priced (I'm not loaded and have to justify it to myself) I'm sure small Vega will do okay pushing a 75hz 3440x1440 panel so I'll make do with that for now and give it 12 to 18 months like I've had to do this time with the Fury pro. Mind you there's no guarantee that Nvidia won't change there policy regarding freesync at some point, They could still sell both by marketng G-sync as the premium option with wider working ranges and faster panels. The option to choose from both camps with freesync panels would be great. Or monitors that support both Gsync and freesync, I'd jump on one of those straight away.
Watch them flip.
Its "Alreet Guys"
He's no a geordie either, he's saying 'Awright Guys', and does everyone south of Scotland think we don't laugh at their accents?
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Seriously? Linus tested an M6000. Lets start with the P6000 which has 3840 cuda cores up to 432GB/s of bandwidth compared to a Titan Z which has 3584 cuda cores and 480GB/s of bandwidth. The P6000 is faster as shown above.
The M6000, has GDDR5, lets start with that, not HBM2, not HBM1, not gddr5x but GDDR5, it has 317GB/s of bandwidth. Then lets move on to, an M6000 has only 3072 cuda cores, oh and a 1140Mhz boost clock compared to the Titan X's 1530Mhz boost clocks.
https://www.youtube.com/watch?v=LC_sx6A5Wko
He's saying the 15% lower cores, the over 30% less bandwidth and almost 30% lower clocks of the M6000, left it a full 1fps behind the Titan X in the Crysis benchmark. But that rumoured massive GDDR5x bandwidth efficiency problem probably isn't responsible for that... well at least Titan X owners will hope.
So I think saying HBM is a bit of nothing based on those using that 'review' of a 3072 cored GDDR5 card is completely insane.
Well, I haven't seen a single comparison of P100 vs Pascal Titan, but lets not forget that both of these cards are 250W. The P100 has an extra 1792 FP64 cores which you know, use power, that is just about half the number of cuda cores over again. Titan X boost clocks which afaik are hit pretty often is 1530Mhz, the P100 boost clock is only 1300Mhz, leaving the Titan X 18% faster ignoring everything else. But it's also extremely unlikely that with all the extra power drain that a P100 will maintain it's boost clock as well as the Titan X does, but again even if it does the Titan X is 18% higher clocked.
For the record there are some latency numbers finally being talked about for HBM and it's not unfavourable compared to GDDR5, in fact, the latency is better for write, about the same for read. GDDR5x I haven't seen any official latency numbers but doubling the prefetch to double the speed would normally significantly increase latency.
Last edited:
Half the stuff the guy says has absolutely zero substance what so ever, the other half is provable wrong fabrciations.
I agree that it will be very disappointing if AMD after nearly a year cannot match the Titan XP using HBM2, new architecture, a more mature process and what seems like a bigger die. Other than that what are saying makes no sense.
Are you suggesting people won't bother looking at the final products benchmarks and reviews, but somehow have noticed that volta image (which hardly anyone noticed until someone did a image grab on a news site) and based their buying decision on that?
Which is obviously why amd and nvidia send him cards to review... right?
I like that accent ^^^^
AMD GTX 1070< for <£350 is just what my Free-Sync Screen is looking for.
So happy I jumped on one of Gibbo's specials and got the 1070 for £350 - had a feeling the price wasn't going anywhere but up any time soon :S and there is no way I'd pay any more for a 1070 for what it is.
@ Roff i'm not paying £100+ more for the same screen to get Nvidia's Adaptive Sync, take that into account Nvidia are way overpriced.
Only because a top on the line AMD card only being GTX 1080'ish means no cheaper than Pascal-Titan-X GTX 1080TI....
You wouldn't buy AMD if it was more Pascal-Titan-X than 1080.
Those who would buy one are actually quite happy as long as the price it right.
Only because a top on the line AMD card only being GTX 1080'ish means no cheaper than Pascal-Titan-X GTX 1080TI....
You wouldn't buy AMD if it was more Pascal-Titan-X than 1080.
Those who would buy one are actually quite happy as long as the price it right.
Last edited:
They send cards to anyone they want, I don't see what that has to do with anything.
Its "Alreet Guys"
Anyway, this guy has been accused of being an AMD fanboy, especially around here, i don't think so...
Once again its just good solid info, GTX 1080+ 10% is fine by me, it means the cut down Vega will probably be GTX 1070 + 10% and not hugely expensive.
Bring it
Its an issue with the video format for a discursive monologue. He fairly often levels criticism at AMD but it it isn't necessarily front-loaded in the intro or title, and unlike in an article format you cant jump to subsections or conclusion page.
It falls under speculation sure but his inferences seem rooted in reasonable reasoning. He often highlights his assumptions.
But he doesn't highlight his outright lies that take 2 minutes on google to show as false.
- Status