Has AMD Reduced RX Vega Supply & Prioritized Frontier Edition?

FP64 performance isn't nearly as important for compute as people make it. Most scientists use it because they don't know or care whether or not they need it.

I have access to 4 p100s, 1 1080 Ti, 1 1080 and 1 980 Ti. FP64 is nice on the p100s because I can use mixed precision for even more performance (there are 50% more ALUs if you use both FP32 and FP64) but otherwise I run in single precision.

What I find strange is that earlier cards had such strong FP64 performance to begin with. The reason we see 1 per 32 these days is to maintain compliance with libraries like OpenCL and CUDA etc.


It's got to be used in some cases where the precision is needed, as there's even higher precision formats than 64 bit, such as 128 bit decimal and even 256 bit decimal if you really want to be insanely precise.


https://en.wikipedia.org/wiki/Double-precision_floating-point_format
 
I don't know. I don't think the OEM's or AMD benefit from the current high prices. The re-sellers, retailers and distributors do by doing markup at the last moment.

The time-lines for production, sales, shipping etc etc etc operate in the region of many many month in advance. Cards arriving today, where probably ordered, built and purchased way before any of the pricing went insane.

Not to mention, the OEM's won't bite. The crypto market is too fast and um-predictable for them to adapt or bother trying to engage. At least for this generation.



In any case, not having a refresh of Vega this year isn't such a big deal, since the 12 nm process is only slightly smaller than the 14 nm that Vega currently uses, so the transistor budget can't go up much more anyhow and according to AMD, the 12 nm process has a 10% power saving over the 14nm one when it's used on Ryzen, so clocks on the latter can go up by the same percentage while staying at the same TDP......Could we see the upcoming Ryzen refresh actually hit 4.4~4.5 Ghz on the overclock front, time will tell.



Things only get more exciting for both GPU makers with the 7nm process, as they both can pack significantly more GPU hardware and keep power at the same levels as today......it's the fab process that may allow hardware twice as fast as today's fastest.
 
I'm looking more in the direction of thread ripper using the updated 12nm ryzen cores, but this time keep 3 of the 4 dies enabled, resulting in a 24 core / 48 thread processor, and really outperforms Intel's 7980x 18 core / 36 thread in multi threading scenarios for less money......Not hard when Intel's chip costs
2000$....:p
 
Last edited:
FP64 performance isn't nearly as important for compute as people make it. Most scientists use it because they don't know or care whether or not they need it.


This couldn't be more further from the truth and a gross negligence of how critical (and long) high level simulations can be.

Modeling anything beyond Newtonian physics is going to require a level of precision that can either make or break the said research in question. You certainly don't want your engine propulsion design to fail because your simulations failed to accurately model minute perturbations in fluid dynamics for example.

Nor do you want your model of the formation and evolution of galactic superclusters and its gravitational effects dismissed outright due to a lack of precision. Cosmological simulations are notorious for taking weeks, if not months, to complete for the level of precision they require.
 
Are you referring to me .... :D



Vega56 + 1080p 60hz monitor :D


Well eventually, games do catch up in terms of GPU requirements as we're still no where near photo realism caliber graphics, and if the idea is to keep the card for as long as possible, since PC gaming is a hobby and not the most important financial responsibility one has, going overkill now for 1080p means having the GPU hardware to last several years.


Even as games eventually become more demanding over time, i tend to punish my own GPU hardware and further shorten it's potential life expectancy by also moving up on the native resolution front which going by what's been announced at CES when it comes to displays, seems to be 21/9 aspect ratio displays with a 5120*2160 resolution.....It's 5.5x harder to render each frame than 1080p on pixel fill rate alone ( just over 11 million pixels every frame rendered ).



This whole crypto currency mining threw everyone off guard by having GPU's cost nearly twice as much as they normally do, and who knows how long this situation will last but one thing is for sure, no one expected paying more for their GPU than they do for a high end CPU, that's for sure.


Badsykes wallet is very happy right now, given he bought his Vega when it was actually affordable....:D :up:
 
We're screwed anywhere how we look at it, as the release of a gaming Volta is entirely optional on Nvidia's end, and in the likely release of said card it's likely that it'll be better than Pascal at mining, so miners will snap them up ASAP and drive the price up big time.....People who want them for gaming will pay thru the nose, not the usual ~700$ introduction price.


Same applies to a Vega refresh at 12nm if AMD releases one......Price tag is likely to be much higher than people want because of mining.
 
Anyhow, It seems that display I have in mind will be out in march ( right around my birthday ), and has a 1500$ price tag to it:


https://www.youtube.com/watch?v=pDM_TbTg3r8



Welcome to the land beyond 4K, and i know the Vega's will hate my guts for increasing their workload even more when it comes to gaming, but that's the kind of guy i am.....:lol:
 
I feel the end of this thread is near too.

Well, there isn't much to talk about that fits the topic of the thread after the first few posts, hence why we have talked about Samsung mining cards, Monitors, etc. Only a small portion has to do with AMD reducing RX Supply and Prioritized Frontier Edition.
 
Well, there isn't much to talk about that fits the topic of the thread after the first few posts, hence why we have talked about Samsung mining cards, Monitors, etc. Only a small portion has to do with AMD reducing RX Supply and Prioritized Frontier Edition.



True, but i just checked far cry 5's system requirements, which is due to be released in march and the 4K 60 Fps requirements are quite high but will put a smile on Vega owners:


http://www.guru3d.com/news-story/ubisoft-far-cry-5-pc-system-requirements.html



Seems that multi GPU is the only way to pull it off, and to run it that way using Nvidia's hardware requires no less than a pair of GTX1080 in SLI, but on AMD's side they can get by with a pair of Vega 56's in crossfire which as we all know with existing games, are not even close to the 1080's level's of performance or it's direct competitor.....That's the job of the 1070 cards.


Here we are worrying about a refresh part not being in AMD's plans, when we may be in for more surprises like this in future game releases with the existing Vega GPU.....:D
 
True, but i just checked far cry 5's system requirements, which is due to be released in march and the 4K 60 Fps requirements are quite high but will put a smile on Vega owners:


http://www.guru3d.com/news-story/ubisoft-far-cry-5-pc-system-requirements.html



Seems that multi GPU is the only way to pull it off, and to run it that way using Nvidia's hardware requires no less than a pair of GTX1080 in SLI, but on AMD's side they can get by with a pair of Vega 56's in crossfire which as we all know with existing games, are not even close to the 1080's level's of performance or it's direct competitor.....That's the job of the 1070 cards.


Here we are worrying about a refresh part not being in AMD's plans, when we may be in for more surprises like this in future game releases with the existing Vega GPU.....:D

How dare we own a Vega, not requiring us to purchase a new card to run Far Cry 5 and future games properly, we are sinners.. LOL :D
 
How dare we own a Vega, not requiring us to purchase a new card to run Far Cry 5 and future games properly, we are sinners.. LOL :D


Or in this case, a pair of Vega 64 liquid editions and always running at their native turbo speeds of 1.75Ghz.......Once I have the 5120*2160 screen I linked above, farcry 5 should look and play quite nicely indeed.


Not bad for the "failure" that is Vega.....:p
 
Or in this case, a pair of Vega 64 liquid editions and always running at their native turbo speeds of 1.75Ghz.......Once I have the 5120*2160 screen I linked above, farcry 5 should look and play quite nicely indeed.


Not bad for the "failure" that is Vega.....:p

That's just it, We don't consider Vega a failure. Each to their own I guess. :D
 
enough already, exposed consider this your final warning, if another thread gets closed where your the central figure again, well you can do the math on that one.
 
enough already, exposed consider this your final warning, if another thread gets closed where your the central figure again, well you can do the math on that one.

Who made me the central figure in the last page or so? I made a comment regarding FP computing, these two decided it was better to blatantly deviate and make personal comments, as you can see in their posts. I don't mind the warning as long as it's universally applied.
 
but in all seriousness everyone needs to a deep breath and relax. Use the ignore function if civility and mutual respect isnt an option. Respect others opinions and debate like you would as IF YOU WERE FACE TO FACE IN REAL LIFE
 
but in all seriousness everyone needs to a deep breath and relax. Use the ignore function if civility and mutual respect isnt an option. Respect others opinions and debate like you would as IF YOU WERE FACE TO FACE IN REAL LIFE

If we were face to face in real life Higgy.





















I'd give you a hug, but not a Rage3D hug because that'd be gay...
 
Back
Top