They doubled the ROPs to 128

 

The specs for Vega 56 in the Code are wrong. Also it is probably best to have five different entries for Vega. One for 56, 64, Nano, Frontier edition with 16GB, and then VII. Do you want the air and liquid card specs separate too?

 

The specs and code is not correct.

 
Hold your horses! I'm in the process of updating the post, will probably take a few days to get it up to par with the RTX 2080 Ti thread.
 

 

Too pricey for me given current prices, let's hope they have a lower VRAM variant at a more reasonable price. I play games, not compute so I'd be looking for gaming performance. Was really hoping to buy one as well.

 

Idk, you may find this interesting:
https://www.techpowerup.com/gpu-specs/radeon-vii.c3358
https://www.techpowerup.com/gpu-specs/radeon-instinct-mi50.c3335

 

I've seen rumored specs showing that VII will have 2x the ROPs at 128, however this is based on the Instinct MI50 correct? That card is limited to 64ROPs.

 

Lisa Su said 128 ROPs at CES yesterday.

https://www.anandtech.com/show/13832/amd-radeon-vii-high-end-7nm-february-7th-for-699

 

It's an interesting product, but at this point I refuse to buy any graphics card that is more than ~ $350.

 

What is the basis of the TimeSpy GPU Scores in the OP? They seem a bit too high AFAIK.

 

 
Results are taken directly from the 3DMark Online Results Browser, the average of a dozen results for each card, taken below the top to exclude the water cooled and LN2 results. So they are overclocked results.

We do know from multiple official AMD performance slides that the VII is very close to the RTX 2080, if not faster in DX12, which Time Spy runs (and there is no reason to doubt the numbers). So putting it at at 12000 (without a specific number in text) for the time being is accurate enough. The point is to show the flagships performance from AMD and Nvidia, how they compete with each other.

I found it interesting that it previously took AMD a year or more to release a direct competitor card to X card from NVIDIA, this time though the difference is only 4 months.
 

 

I wish I do not become a member of this club, but I am quite unsure I will be able to manage it.

 

will be getting a V7 as soon as becomes available for pre order.

 

So, why is CEO of nVidia saying that the Vega VII performance is laughable, and lousy?

It looks strong to me thus far.

I did however buy a Titan XP Galactic empire, being impatient for the same price.

I may still grab one or two of these. For testing, price inflation, resale purposes.

 

So, why is CEO of nVidia saying that the Vega VII performance is laughable, and lousy?

Perhaps he's cranky because his company's stock lost 50% of its value in the past 2 months and has a class action lawsuit against it for misleading investors.

Seriously though, Vega VII is not the failure the community is making it out to be, at least on paper. In gaming it's good(not great), but in productivity it's a beast. Those OpenCL gains are gigantic and the card will certainly find a place in the systems of those users, especially with 16gb of HBM2. Also, Ryan Smith from Anandtech is reporting that FP64 is fully intact which means the card has nearly the double precision performance at the Titan V at 1/4 the cost.

 

Even as an avid AMD user I can see his take. It's new process vs old, IIRC 7nm vs 12nm. Basically new process allowed clocks gains, etc, keeping power usage same as last gen. Nvidia still more efficient even on older process.

Initially I did not go VEGA at launch, I felt it was overpriced and lacking performance. I tried a GTX 1080, yeah thing OC'd like a bitach, hit my loop way less than VEGA IMO, but not having VRR (as was not gonna pay G-Sync tax) was a big PITA, the driver panel was reminiscent of when I owned a GTX 8800 >.< .

So later when I saw a deal on VEGA, I sold GTX 1080. Yeah it consumes more power, hits loop harder, but I'm enjoying the use of FreeSync and better driver panel.

I reckon nVidia allowing FreeSync monitors is a grab at owners that may reel G-Sync tax, it could have kept a defector like me away from VEGA at that time, as I had an ASUS MG278Q.

 

I’m not that impressed with nVidia’s latest generation turning. Sure, it is fast but so was the last gen.

If you go look at the RTX Titan, the fastest video card on the block, but it cost $2,500 bucks. On a custom loop, with water block, and overclocked to the max it is 22% faster than a air cooled overclocked Titan XP.

 

That card is targeted at researchers who need more memory not ECC memory. It even supports NVLink memory pooling for an effective 48GBs of usable memory which only Quadros/Teslas support. That card isn't targeted at gamers and at the price it isn't worth it for gaming. If Vega II actually has full double precision performance it's a bargain and I'll pick one up.

 

Count me in for 2 once some blocks come out! Although you should update the ROPS, 128 was confirmed a mistake said by someone & the media went with it, they've since corrected the statement, it's actually 64.

https://www.extremetech.com/gaming/...h.com/gaming/283649-the-amd-radeon-viis-core-configuration-has-been-misreported

 

(2) Vega VII’s will be awesome for sure. A ton of power usage never bothered me much.


Anyone know when freesync 2 will be true 30-144Hz? Or if it already is? I’m seeing a lot of misinformation on this.

G-Sync was awesome with 4K, because when the frame rate dipped to 30’s it really felt incredible. It felt like 45 FPS. And it wasn’t wavvy feeling like 30-38fps range is.

But freesync is so much cheaper. I wish they could fix this. AMD is awesome, and there going in the right direction regardless.

It’s just 48-60Hz isn’t good enough.

 

This card is tempting, i hope we will see actual reviews very soon, for the price though i don't know, i wish it was between the RTX2080 and RTX2080Ti at least, leaning closer to the 2080ti. That would've made my decision much easier... But i guess that's wishful thinking. The aesthetic on the other hand is dope.. clean and classy. You guys think it'll be their flagship card for the whole year?

 

It’s tough making a choice on which GPU for everyday use. Honestly though, RTX2080 and Vega VII are going to be the high end gpu performance standard for the next 2 years or longer.

That’s why it was a pretty easy choice for me to grab a Titan XP. It can beat a RTX2080 once it is overclocked real good, and it has more vram all of this for $700 bucks, plus nvidia unlocked professional performance capabilities for it.

But, I will be buying a Vega VII for testing purposes. And to support AMD. I’m getting a Ryzen 2 as soon as there available.

 

Eh if you get this GPU be sure to run Milkyway@Home or some other FP64 application, maybe something with over 11GB VRAM use. Not much value otherwise. The misreporting of 128 ROPs is a disappointment. GTX 1080 TI has 88 ROPs and so does RTX 2080 Ti, so this is essentially surpassing GTX 1080 and RTX 2080 only. With any luck it will match GTX Titan X Pascal in pro workloads that don't use CUDA.

Also judging by the fact that only VEGA FE had pro drivers, I don't think it'll exceed VEGA64 by over 27% in pro viewports such as Solidworks or Siemens NX.

CATIA isn't overly dependent on pro drivers for AMD cards. For Nvidia cards it's a different story. Also for Blender: that's an ideal use case for the VRAM.

See: https://www.anandtech.com/show/12170/nvidia-titan-v-preview-titanomachy/6 , https://www.gamersnexus.net/guides/...guides/3399-best-workstation-gpus-2018-for-adobe-premiere-autocad-vray-and-more , https://techgage.com/article/specviewperf-13-viewport-performance/, https://www.notebookcheck.net/GeFor...top-vs-NVIDIA-Titan-RTX-vs-GeForce-RTX-2060-Laptop_9526_9651_9581.247598.0.html


Going off Gamersnexus' review, +27% would make it still slower than GTX 1080 Ti in Maya although for Solidworks it'd be faster than GTX Titan XP provided it does scale +27%. +27% in Autocad would make it semi competitive , whereas Creo / NX both require pro drivers on AMD. Of course scaling isn't linear since these benchmarks are also partly CPU bound so I don't expect it to make any significant improvements over VEGA 64 and VEGA FE.




https://hwbot.org/benchmark/specviewperf_12_catia/
201.88 FPS

Cautilus Titan V @ 1980/1026MHz <-- R7 2700X @ 4.35GHz might be holding this back since it is a quicker GPU than GTX 1080
149.06 FPS

MrGenius Radeon RX Vega 64 @ 1690/1175MHz <-- Ivy Bridge @ 5.3GHz might be holding this back?
147.2 FPS

Baddemichl Quadro P5000 @ 1657.5/1499.8MHz <--- GTX 1080 , pro drivers
139.33 FPS

rambustibi Radeon Vega Frontier Edition @ 1377/800MHz <-- Threadripper is suboptimal CPU for this, but it was running 3.99GHz
128.68 FPS

randyenergy GeForce GTX 1080 Ti @ MHz <--- was running stripped OS on i5 haswell @ 4.9GHz
115.65 FPS

CjMitsuki GeForce GTX 1080 Ti @ 2088/1626MHz <-- GTX 1080 Ti overclocked to the wall on ambient pretty much , 4.55GHz R7 2700X



Benchmark download: ftp://ftp.spec.org/dist/gpc/opc/viewperf/SPECviewperf12_1_1.zip
(The listed cards above my $400 P2000 , which happens to use 75W)

 

I'm tempted to buy one also if not just to see how far I could push the core clock. Anyone know how much the 16GB HBM would contribute to the supposed 300w TDP?

 

I dunno, seems like the only benefit over vega 64 is a few 100mhz over vega 64 and double vram.

 

And the highest VRAMspeed of any card (1TB/s). Not insignificant. This will imho give it more longevity than a 2080.

Debating 1 or 2 on my 8k monitor.

 

price Italy- Spagna
Bye

 

Is there no place to pre-order this card in US?

 

No pre-orders. Just like the Vega release.

 

first test

 

those are AMD's #s since CES

 

XFX Radeon VII just leaked ahead of schedule on newegg US. Got my order in! Says it will ship tomorrow!

https://www.newegg.com/Product/Prod...=N82E16814150820&Description=radeon vii&cm_re=radeon_vii-_-14-150-820-_-Product

 

How did you even find it? Search for Radeon VII doesn't show any results of this card.

 

 

It looks like a very interesting card but for me I may hold on to my Radeon Vega Frontier Edition 16GB HBM2 card. Mainly because I am working with Unity + C# Programming plus casual gaming. Now question is, I though Ray tracing for AMD was going to be software base and if, so will the Vega FE get an updated driver?