• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Vega countdown has appeared

Even marketing is outsourced.

Domain name:
VE.GA is Active

Owner contact:
Organization: Brand & Deliver HQ Ltd
Name: Ben Gallop
Address: 5-6 Mallow Street
Zipcode: EC1Y 8RQ
City: London
State: England
Country: United Kingdom
Phone: +44-2074909200
Fax: N/A
E-mail: nick@brandanddeliver.co.uk

Domain registered: 11/03/2016
Record will expire on: 11/03/2017
Record maintained by: My GA Domain Registry
 
Now just a google search of talking bad about competition apparently this adverting firm doesn't know anything marketing or AMD asked them to do something they normal don't do

https://www.google.com/search?q=nev...safe=off&q=talking+bad+about+your+competition

One thing to show you are better with stats and data, another to talk crap even if you can show data and stats at that point they don't need to say anything cause stats and data take over.
 
So far it looks like the #betterRed marketing campaign is succeeding. :thumbs:

It looks like AMD might know something about Volta that most don't. Maybe it's a dud. We know it's already seen a big delay, i wonder what went wrong.

At any rate, it looks like the normal cadence. Even if at a minimum Vega beats 1080ti, then that gives AMD top spot for the better part of 2017? Will NV rush out Volta even if it is having issues?
 
So far it looks like the #betterRed marketing campaign is succeeding. :thumbs:

It looks like AMD might know something about Volta that most don't. Maybe it's a dud. We know it's already seen a big delay, i wonder what went wrong.

At any rate, it looks like the normal cadence. Even if at a minimum Vega beats 1080ti, then that gives AMD top spot for the better part of 2017? Will NV rush out Volta even if it is having issues?


Outside of what nV has shown so far AMD has no clue what Volta is, even nV doesn't have final knowledge of Volta at this point, and you think AMD has any sold info on Volta?

That is a bit whimsical at this point. There is no "rushing" of products, even if nV needs to push Volta forward, the most they can do is a quarter, that still puts it at end of 2017....... So no that doesn't make sense either.

And this is not normal cadence, really, a Performance, Enthusiast product coming out 4 quarters after their competition? This is the LONGEST gap between launches for AMD and nV EVER! Previous longest was Fiji which was 3 quarters, before that, r600 and fx series 2 quarters (Fermi could be slotted in here too).
 
Outside of what nV has shown so far AMD has no clue what Volta is, even nV doesn't have final knowledge of Volta at this point, and you think AMD has any sold info on Volta?

That is a bit whimsical at this point. There is no "rushing" of products, even if nV needs to push Volta forward, the most they can do is a quarter, that still puts it at end of 2017....... So no that doesn't make sense either.

And this is not normal cadence, really, a Performance, Enthusiast product coming out 4 quarters after their competition? This is the LONGEST gap between launches for AMD and nV EVER! Previous longest was Fiji which was 3 quarters, before that, r600 and fx series 2 quarters (Fermi could be slotted in here too).

Yeah it's normal cadence. And of course AMD knows what's up with Volta (or down as the case may be), even if Nv has changed their roadmap as often as a baby's diaper. As far as Vega goes, i think intel dumping nvidia says a lot.

What's really going to be interesting is Raven Ridge with Vega, Zen and Infinity fabric all rolled into one nice HSA package.
 
Yeah it's normal cadence. And of course AMD knows what's up with Volta (or down as the case may be), even if Nv has changed their roadmap as often as a baby's diaper. As far as Vega goes, i think intel dumping nvidia says a lot.

Intel dumping on nV tells us one thing they are afraid of the HPC space and what nV has done as an ecosystem

What's really going to be interesting is Raven Ridge with Vega, Zen and Infinity fabric all rolled into one nice HSA package.

And we have no clue how the uptake that will be, mainly because of the software side, please look at the many threads on HSA and HPC about this, AMD needs to work very hard to catch up on the software side.

This is atypical cadence releasing high end products this late in the game, cause people that buy high end products tend to do so early on since they have the money to buy one in the first place, their expendable cash on hobbies tend to be higher then your mainstream and value buyers. Waiting this long for a competitive product, it has to be better by a margin (in all metrics) large enough for a performance and enthusiast is forced to upgrade. As we have seen any time cards have been delayed more than a quarter then the competition, they always were at the short end in one form or another. FX series, r600, Fiji (not performance, but power usage). Some part of their metrics hard to deny the difference.
 
Last edited:
Does anyone expect vega to keep up or beat 1080 Ti overall? I think they'd be lucky to reach 1080 levels based on 480 performance. AMD probably knows as much as we do about Volta which is next to nothing, this is just marketing fluff.
 
Does anyone expect vega to keep up or beat 1080 Ti overall? I think they'd be lucky to reach 1080 levels based on 480 performance. AMD probably knows as much as we do about Volta which is next to nothing, this is just marketing fluff.
Um not a single person with a clue expects this to touch the 1080 ti. Anyone that has followed this in the least knows AMD is just trying to match the 1080. Even their own best case scenario using Vulkan in Doom shows it only barely beat the 1080. Overall I am guessing it will be 5-10% slower than the 1080 in reviews.
 
Intel dumping on nV tells us one thing they are afraid of the HPC space and what nV has done as an ecosystem.

I think you missed that developers are eager for an open ecosystem, not a walled up proprietary ecosystem. It might tell you that intel is afraid of nv, but it tells me that nv is not a great partner to work with, and AMD's hardware is the better choice.
And we have no clue how the uptake that will be, mainly because of the software side, please look at the many threads on HSA and HPC about this, AMD needs to work very hard to catch up on the software side.

And they have been and have most of the software in place, as have their partners. Intel is now talking about a complete redesign of their x86 architecture (timing seems...telling) to drop a bunch of instructions, sometime around the 2020-2021 timeframe from rumors so far. IMO this is their answer to Moore's law slowing drastically over time, which is what HSA is here to address already.

This is atypical cadence releasing high end products this late in the game, cause people that buy high end products tend to do so early on since they have the money to buy one in the first place, their expendable cash on hobbies tend to be higher then your mainstream and value buyers. Waiting this long for a competitive product, it has to be better by a margin (in all metrics) large enough for a performance and enthusiast is forced to upgrade. As we have seen any time cards have been delayed more than a quarter then the competition, they always were at the short end in one fperformance, but power usage). Some part of their metrics hard to deny the difference.

This is normal cadence, AMD has Polaris on the market ahead of Pascal. The only difference is they are skipping the 14nm port from a 20nm design (aka Polaris) at the high end, and releasing Vega, the true 14nm design. IMO.
 
think you missed that developers are eager for an open ecosystem, not a walled up proprietary ecosystem. It might tell you that intel is afraid of nv, but it tells me that nv is not a great partner to work with, and AMD's hardware is the better choice.


No I didn't, the enthusiasm is there, but the capability of AMD hardware via libraries and extensions and the implications on performance because of those limitations is rather large. I suggest you look it up on AMD's own website about those features, they have been "planning" on implementation of those features for the past year, still haven't done em yet.

Intel got nothing from nV, they were court order to pay nV so....... yeah there is bad blood there. But if you see how Intel is trying to stop nV in HPC's its obvious. Intel has problems against nV in HPC markets, you think AMD would fair better?
And they have been and have most of the software in place, as have their partners. Intel is now talking about a complete redesign of their x86 architecture (timing seems...telling) to drop a bunch of instructions, sometime around the 2020-2021 timeframe from rumors so far. IMO this is their answer to Moore's law slowing drastically over time, which is what HSA is here to address already.

Look above, you don't know what you are talking about, its not just instructions, there are full libraries that aren't fully compatible yet.


This is normal cadence, AMD has Polaris on the market ahead of Pascal. The only difference is they are skipping the 14nm port from a 20nm design (aka Polaris) at the high end, and releasing Vega, the true 14nm design. IMO.

Yeah your opinion is quite wrong lol. What? what are you saying, the gp104's cards were one month in front of Polaris lol, dude, what your smoking must be damn good. AMD THOUGHT they had a huge lead, then Pascal, gp 104 was released in May of last year, and Polaris came out in June of last year......
Do I need to link this or .......

Now if you are talking about gp106, Polaris had a week or two advantage, hardly an advantage, added to that, supply for Polaris was horrid for a month or two.

Sorry man don't know what you are talking about.
 
Looks like Roy poking the sleeping bear again. Forgot where it was posted last year but the rumor was Volta was taped out already and could be out by May, just sitting on it refining it. But with lack of competition Nvidia could refine Pascal and push out another series before Volta. Even if AMD Beats them good with Vega by what? March/April time frame for actual card release, Nvidia could switch gears back to Volta for a September release.

Blah. We will be back to the every six to nine months one trumps the other. Then they will be found colluding on price fixes again blah blah blah blah... pick your poison, red pill green pill. Can we get a third party already???
 
Looks like Roy poking the sleeping bear again. Forgot where it was posted last year but the rumor was Volta was taped out already and could be out by May, just sitting on it refining it. But with lack of competition Nvidia could refine Pascal and push out another series before Volta. Even if AMD Beats them good with Vega by what? March/April time frame for actual card release, Nvidia could switch gears back to Volta for a September release.

Blah. We will be back to the every six to nine months one trumps the other. Then they will be found colluding on price fixes again blah blah blah blah... pick your poison, red pill green pill. Can we get a third party already???


Hmm I don't think Volta is taped out yet, a bit too earlier, I think tape out will be around June for end of year or early next year launch
 
No I didn't, the enthusiasm is there, but the capability of AMD hardware via libraries and extensions and the implications on performance because of those limitations is rather large. I suggest you look it up on AMD's own website about those features, they have been "planning" on implementation of those features for the past year, still haven't done em yet.

Intel got nothing from nV, they were court order to pay nV so....... yeah there is bad blood there. But if you see how Intel is trying to stop nV in HPC's its obvious. Intel has problems against nV in HPC markets, you think AMD would fair better?


Look above, you don't know what you are talking about, its not just instructions, there are full libraries that aren't fully compatible yet.




Yeah your opinion is quite wrong lol. What? what are you saying, the gp104's cards were one month in front of Polaris lol, dude, what your smoking must be damn good. AMD THOUGHT they had a huge lead, then Pascal, gp 104 was released in May of last year, and Polaris came out in June of last year......
Do I need to link this or .......

Now if you are talking about gp106, Polaris had a week or two advantage, hardly an advantage, added to that, supply for Polaris was horrid for a month or two.

Sorry man don't know what you are talking about.


Yet AMD gained market share so it looks like they have a better clue of the industry than yourself. Vega will be their true 14nm design, while Polaris was the 14nm pilot (which was released to market before Pascal). AMD is confident they'll reach a 50% market share, i'm confident they will also. As for your HPC market, yes i do think the Naples platform with associated accelerators will challenge nvidia in the HPC market.
 
lolamdmarketing.png
 
Yet AMD gained market share so it looks like they have a better clue of the industry than yourself. Vega will be their true 14nm design, while Polaris was the 14nm pilot (which was released to market before Pascal). AMD is confident they'll reach a 50% market share, i'm confident they will also. As for your HPC market, yes i do think the Naples platform with associated accelerators will challenge nvidia in the HPC market.


That wasn't due to Polaris, if you look at marketshare numbers Q2 is when AMD gained marketshare, Q3 they lost some, so Polaris came out In end of Q2 Last week, Q3 tells up Polaris even though AMD volume was up, nV's volume uptake of Pascal was more cause it had to overcome the 70 to30 % disparity of Q2 which it did. Added to this notebook marketshare which AMD pulled off in Q2 increase is still there in Q3 because of getting rid of old AMD chips at a lower margin. So that in favor or AMD still, need to take that out, to figure out Polaris's gain which would impact it more to the negative direction.

So you are either BS'ing yourself to believing AMD's marketing, or just flat out wrong, either way, its wrong.

HPC market is not all about performance, if the software isn't there then there is a problem. the people making software for HPC are scientists, they don't give a shit about the hardware as long as the hardware and software can do what they need. (faster is better) But if the libraries aren't there, they can't do what they need. AMD has had a FLOPS advantage for years now, and this is what is holding them back, software just can't compete yet because of the lack of libraries and extension support.
 
Last edited:
Looks like Roy poking the sleeping bear again. Forgot where it was posted last year but the rumor was Volta was taped out already and could be out by May, just sitting on it refining it. But with lack of competition Nvidia could refine Pascal and push out another series before Volta. Even if AMD Beats them good with Vega by what? March/April time frame for actual card release, Nvidia could switch gears back to Volta for a September release.

Blah. We will be back to the every six to nine months one trumps the other. Then they will be found colluding on price fixes again blah blah blah blah... pick your poison, red pill green pill. Can we get a third party already???

I think nvidia had serious problems with Volta, hence the complete overhaul of their roadmap and they went back to the drawing board on it. This may be what AMD is referring to in this video.
 
That wasn't due to Polaris, if you look at marketshare numbers Q2 is when AMD gained marketshare, Q3 they lost some, so Polaris came out In end of Q2 Last week, Q3 tells up Polaris even though AMD volume was up, nV's Volume uptake was more.

So you are either BS'ing yourself to believing AMD's marketing, or just flat out wrong, either way, its wrong.

AMD attributed much of their market share gains to the 300 series (probably due to AMD's FineWine technology in action. Or Nvidia stale beer technology whichever),and as you say AMD's GPU volume was up in Q3 while sliding meager amounts, all while having 3 new cards on the market while nv had 5. Therefore the Polaris was in higher demand than 1060 and 1050's. At this point I can't decide if you are just swallowing NV's BS marketing, or actually writing it.
 
AMD attributed much of their market share gains to the 300 series (probably due to AMD's FineWine technology in action. Or Nvidia stale beer technology whichever),and as you say AMD's GPU volume was up in Q3 while sliding meager amounts, all while having 3 new cards on the market while nv had 5. Therefore the Polaris was in higher demand than 1060 and 1050's. At this point I can't decide if you are just swallowing NV's BS marketing, or actually writing it.


really already explained everything to you.

Now you are just asking for the links?

I can give you links to everything I have stated. So are you going to sit here and BS me and everyone else on this board?

Please I'll give 45 minutes till the Green Bay, Lions game is done, then I'll post the links.

And you can start with the HPC ones go to google, type in AMD GPU Open HPC features vs Cuda. that will start it off.
 
NPP – Performance Primitives for Image & Video Processing
FFT – Fast Fourier Transforms Library
BLAS – BLAS Library
SPARSE – Sparse Matrix Library

All these libraries are not fully supported (extensions within each library or CUDA has extra features that make things faster in some of them.)

https://wiki.tiker.net/CudaVsOpenCL (this is something that was talked about 3 months ago, fairly recent)

Open CL is what GPUopen is using

Personally, I would like to see OpenCL succeed. It has the right ingredients as a standard--mainly run-time code generation and reasonable support of heterogeneous computing. On top of that, being in a multi-vendor marketplace is a good thing--also for Nvidia, although they might not immediately see it that way.
If I was starting something new, I would likely go with OpenCL, unless I desperately needed one of the proprietary CUDA libraries. --AndreasKloeckner

That being said, I also have to have a separate program to debug syntax in OpenCL. CUDA can be used straight through Visual Studio, and it has intellisense. CUDA can also use variables straight out of code, due to it being code. OpenCL is parsed as a string. The CUDA environment is much more user friendly. OpenCL has more "customizable" options for it, but this just leads to code refactoring between machines. CUDA seems to be able to port much more consistently, and its easier to work with Development Environments with CUDA. Overall, I have done OpenCL for 2 months, CUDA for 2 days, and I have had more success with CUDA.


This is the problem, most HPC users if they are already on CUDA, they aren't going to switch off, because they are comfortable with what they have, now people making new software for HPC, AMD still has to overcome nV's pushing CUDA in colleges, and they are way ahead there.

now lets talk about marketshare

MW_table_q316.PNG


So AMD lost marketshare with Polaris not gained. AMD did not attribute marketshare gains from 3xx series. They have pretty much stopped talking about the last gen products outside of Fiji cards for the past 2 Q's But we can see this because, OEM's like Dell, HP, and Lenovo, picked up 3xx series and Fiji based cards one year after their lunch. Which those cards to do not compete with Pascal based cards very well in most metrics, which also were seen with the increase of AMD's marketshare in Q2.

So can't expound upon your reasonings? Cause I just showed ya everything contrary to what you have stated.
 
Oh boy we are going to have some fireworks next 4 days and until the day we see vega come out. Now on the positive side this looks like a open preview of the architecture since they have a count down.
 
Oh boy we are going to have some fireworks next 4 days and until the day we see vega come out. Now on the positive side this looks like a open preview of the architecture since they have a count down.

I just hope the preview will contain juicy substances rather than some dumb powerpoint slides and graphs like the Polaris preview from same time last year.
 
I just hope the preview will contain juicy substances rather than some dumb powerpoint slides and graphs like the Polaris preview from same time last year.
it should. they already touched on it during the horizon thing with the battlefront "demo". we'll prob get more info and "demos" that they pick to showcase it. and of course the slides...
 
it should. they already touched on it during the horizon thing with the battlefront "demo". we'll prob get more info and "demos" that they pick to showcase it. and of course the slides...


Hopefully ;) they showed Polaris off months before a couple of times before the slide show of it ;)
 
Childish outbursts will get you banned.
 
So far it looks like the #betterRed marketing campaign is succeeding. :thumbs:

It looks like AMD might know something about Volta that most don't. Maybe it's a dud. We know it's already seen a big delay, i wonder what went wrong.

At any rate, it looks like the normal cadence. Even if at a minimum Vega beats 1080ti, then that gives AMD top spot for the better part of 2017? Will NV rush out Volta even if it is having issues?
Some details have sort of come out about Volta performance, and if they end up being true I think AMD is being a bit idiotic to do such marketing and taunting a product that is meant to be another leap like Maxwell was in terms of performance-efficiency.
This is going by some of what has come out regarding the Volta 'Tegra' (samples actually anticipated by Nvidia to be with automobile manufacturers Q3 this year) and also the Volta 'V100' going into the next supercomputers working with IBM for starting deployment around mid year-to-Q3.
Cheers
 
Some details have sort of come out about Volta performance, and if they end up being true I think AMD is being a bit idiotic to do such marketing and taunting a product that is meant to be another leap like Maxwell was in terms of performance-efficiency.
This is going by some of what has come out regarding the Volta 'Tegra' (samples actually anticipated by Nvidia to be with automobile manufacturers Q3 this year) and also the Volta 'V100' going into the next supercomputers working with IBM for starting deployment around mid year-to-Q3.
Cheers
what details(links/quotes/names) and what do cars and ibm super computers have to do with vega/gaming?
 
what details(links/quotes/names) and what do cars and ibm super computers have to do with vega/gaming?
A lot as it is complmentary when considering efficiency-performance and the design focus for modern GPUs is also for HPC/Deep Learning/inferencing, and well the Pascal gaming cards have very good efficiency-performance andwith a stronger GPU-silicon performance envelope than before.
I would need to dig them up, but it relates to the power demand figure of Volta Tegra (Xavier) being much lower than Drive PX2 and with same performance as the full Drive Pascal PX2+GPUs, and the Volta 'V100' based upon the finalised performance report given by Nvidia/IBM to several of the supercomputer projects they are contractually obliged to do-start in 2017.
The figures work out to be like the jump we saw with the creation of Maxwell or Pascal, so IMO rather stupid for AMD to taunt about Volta as they are, would be better if they just ignored it.
Cheers
 
Last edited:
The problem with this whole endeavour is that it leads to another small piece of the puzzle rather then something substantial. The build up might work if they would announce what the topic is. Flashing images of things to come and some graffiti trash talking is not really a good example of building interest for a product.

Nvidia has to be laughing as they sell us gimped cards for 1200 bucks with no competition in sight.
That is the problem for people paying that amount of money for something they can't live without.
You don't have to worry about one thing as it is now AMD won't release cards that expensive either (not counting the pro series).

The amount of money and the brand name behind it says it all really ...
 
And did you see nV outside of performance graphs in slides talk about Polaris? Did they have to? Nope and the 1060 is outselling the rx480....

Chill dude. This style of marketing has been used to great success for years. AMD is the underdog and Nvidia is the incumbent. It makes sense for AMD to compare themselves to Nvidia and for Nvidia ignore them.

Pepsi's done this to great effect. Apple used to do this before they became the juggernaut that they are today. Insurance companies engage in this continuously.

https://en.wikipedia.org/wiki/Comparative_advertising

I for one am hoping an praying to see a little more competition in the high end GPU market, and if AMD can deliver 1080 like speeds at a 1070 price or lower, I'll be over the moon.
 
Chill dude. This style of marketing has been used to great success for years. AMD is the underdog and Nvidia is the incumbent. It makes sense for AMD to compare themselves to Nvidia and for Nvidia ignore them.

Pepsi's done this to great effect. Apple used to do this before they became the juggernaut that they are today. Insurance companies engage in this continuously.

https://en.wikipedia.org/wiki/Comparative_advertising


Comparative advertising is different, comparative adverting always has data backing it up, what AMD did, is they have nothing to back them up yet.

Comparative advertising has been increasingly implemented through the years, and the types of comparative advertising range from comparing a single attribute dimension, comparing an attribute unique to the target and absent in the referent and comparisons involving attributes unique to both brands.[34] The contributing factors to the effectiveness of comparative advertising include believability,[35] which refers to the extent a consumer can rely on the information provided in comparative advertisements, the level of involvement,[36] and the convenience in evaluation,[37] provided by spoon feeding the consumer with information that does not require extra effort in recall.[38]

From your own link. AMD has done neither of this and if they don't show it, its just talking a lot of hot air.

Yes I have worked in advertising as a consultant for FTI, a major Firm that advertising companies (and other industries) hire when they need to get rid of the fat in their company, and have restructured some of the largest adverting firms in the world in NYC, from traditional print and tv marketing, online marketing, and emerging markets.
 
Comparative advertising is different, comparative adverting always has data backing it up, what AMD did, is they have nothing to back them up yet.

The two examples I provided (Pepsi vs Coke & Mac vs PC) are evidence to the contrary. The comparison can be implied or explicit.

In the case of AMD, they are suggesting that they are bringing VR at a lower price than their competition.

In the case of Apple vs PC, Apple claimed it was bringing a cooler product.

In the case of Pepsi vs Coke, Pepsi claimed that it was a younger and cooler alternative.

 
Back
Top