Does it suck for the average gamer? Well, we will have to wait for actual reviews, but, the bang for the buck doesn't appear to be there, unless you do those bundles.
Vega 56 at $400 and 210W TDP (and going by recent AMD offerings, about 90-95% of Vega 64) sounds like an excellent deal. It reminds me of R9 290. I expect good overclocks as well, getting the card to stock Vega 64 performance levels.
If it also supports HDMI 2.1 (and variable refresh rates, same as Display port adaptive sync), it'll be the card to get for many gamers. Though this feature is more likely to debut with Navi.
AMD's RTG needs to have already started design of a post GCN architecture. Its painfully obvious that Vega is a disaster for gaming. I cannot believe that AMD regressed on Fiji and Polaris on real world gaming perf/watt. AMD needs to take a leaf out from Nvidia's playbook. Two separate dies - one for pure gaming and the other for HPC, Machine learning. A post GCN architecture which uses a multi die approach connected with Infinity fabric is probably the future for AMD GPU designs. For RTG the next few years will be very bad. But sometimes its necessary for such a disaster to force a clean sheet design for the future. I cannot believe AMD have failed so hard. Even Bulldozer was able to compete on multi thread workloads but was horrible for single thread performance and perf/watt. Vega is just a worse disaster than Bulldozer. Vega's gaming perf/watt and perf/sq mm is the worst ever for a GCN GPU. After 10 years AMD have delivered another HD 2900XT.
I think they aimed for much higher clock frequency but failed. I guess GCN just doesn't scale.AMD's RTG needs to have already started design of a post GCN architecture. Its painfully obvious that Vega is a disaster for gaming. I cannot believe that AMD regressed on Fiji and Polaris on real world gaming perf/watt. AMD needs to take a leaf out from Nvidia's playbook. Two separate dies - one for pure gaming and the other for HPC, Machine learning. A post GCN architecture which uses a multi die approach connected with Infinity fabric is probably the future for AMD GPU designs. For RTG the next few years will be very bad. But sometimes its necessary for such a disaster to force a clean sheet design for the future. I cannot believe AMD have failed so hard. Even Bulldozer was able to compete on multi thread workloads but was horrible for single thread performance and perf/watt. Vega is just a worse disaster than Bulldozer. Vega's gaming perf/watt and perf/sq mm is the worst ever for a GCN GPU. After 10 years AMD have delivered another HD 2900XT.
AMD's RTG needs to have already started design of a post GCN architecture. Its painfully obvious that Vega is a disaster for gaming. I cannot believe that AMD regressed on Fiji and Polaris on real world gaming perf/watt. AMD needs to take a leaf out from Nvidia's playbook. Two separate dies - one for pure gaming and the other for HPC, Machine learning. A post GCN architecture which uses a multi die approach connected with Infinity fabric is probably the future for AMD GPU designs. For RTG the next few years will be very bad. But sometimes its necessary for such a disaster to force a clean sheet design for the future. I cannot believe AMD have failed so hard. Even Bulldozer was able to compete on multi thread workloads but was horrible for single thread performance and perf/watt. Vega is just a worse disaster than Bulldozer. Vega's gaming perf/watt and perf/sq mm is the worst ever for a GCN GPU. After 10 years AMD have delivered another HD 2900XT.
It's not rocket science. Vega is the result of bad decisions. How the hell were they able to do same performance/clock as FIji with a GPU with 2x the transistors. Fiji has 8.9 mil and Vega has 15-18mil. Yet they perform the same at the same clock.
It's not rocket science. Vega is the result of bad decisions. How the hell were they able to do same performance/clock as FIji with a GPU with 2x the transistors. Fiji has 8.9 mil and Vega has 15-18mil. Yet they perform the same at the same clock.
Hmm, you're right. I should have researched more. Still, that's no excuse.12,5B
12,5B and you are comparing a mature product with a newborn one with several features not enabled yet.
.
I think the time for falling back on that argument, has passed.
AMD has put their cards on the table, and we now know they have been bluffing all along. They had nothing and they knew it, and a lot of us suspected it, but now we know.
Vega is train wreck.
True. AMD have zero credibility in the GPU business. It will take multiple generations of strong execution and many years for them to build credibility and trust among enthusiast PC gamers. This decade belongs to Nvidia - they are an execution juggernaut.As a professional/compute card, Vega isn't too bad. Radeon WX 9100 should sell decently; at $2,199 it's only a bit more expensive than Quadro P5000, but far more powerful.
The problem isn't that AMD is not competing in the high-end gaming segment; writing that off is a plausible business decision. The problem is that their marketing team was so flagrantly dishonest about it. I've lost a lot of respect for AMD over this fiasco. Any gamer who bought into the hype has been stuck with lesser products for months "waiting for Vega" when they could have had a superior Nvidia card at a price no higher than Vega will now cost. Worse, prices have become inflated from the mining craze, with the GTX 1070 almost unavailable and the GTX 1080 temporarily back up in price, so anyone who waited is likely to now incur a real financial loss unless they are willing to wait longer.
False. According to AMD itself, and it is clearly stated on several indipendent sites, i.e. tile based rasterization is not enabled in the drivers. Enhanced output for the geometry is not fully functional. Primitive shaders are not used by applications. Also, tier 3 of DX12 features need to be used by application.
So, there is some untapped potential still present. How much, it's difficult to say. I'm far from hinting to "miracolous drivers" but a +10% compared to actual situation, averaging across the board, seems feasible. Moreover, all the performance figures are referred to the Vanilla - air cooled- version. AIO board will quite probably be faster even if price premium seems not so good on that one.
How can you say what is enabled in the RX Vega drivers, when there are none to test?
AMD have finally revealed their own RX Vega performance comparisons. All companies cherry pick these to show their own products in the best possible light, and even then is clear they don't beat GTX 1080.
AMD themselves say it will be "trade blows" GTX 1080.
16 months after GTX 1080, the release a card that trades blows with it, but requires an extra 100 watts, and a Titan sized die.
It's a train wreck.
Speaking of Fiji, there’s been some question over whether the already shipping Vega FE cards had AMD’s Draw Steam Binning Rasterizer enabled, which is one of the Vega architecture’s new features. The short answer is that no, the DSBR is not enabled in Vega FE’s current drivers. Whereas we have been told to expect it with the RX Vega launch. AMD is being careful not to make too many promises here – the performance and power impact of the DSBR vary wildly with the software used – but it means that the RX Vega will have a bit more going on than the Vega FE at launch.
Exactly. Maybe sometimes reading the articles of this site before posting on its forum can help.He can say it because Ryan Smith reported it on the main page of AT.
Exactly. Maybe sometimes reading the articles of this site before posting on its forum can help.
Also, there are comments from other people as well, like sebbbi in the B3D forum.