Editorial AMD Didn't Get the R9 Fury X Wrong, but NVIDIA Got its GTX 980 Ti Right

[revanchrist]

Kind of disappointed by Fury X's performance. Was expecting more. Now i can only wait for Pascal. i know Arctic Islands is coming @ Q4 2016. That's too far away. Now the only wildcard is Pascal, hoping they'll launch it at 1H 2016.

 

[MrGenius]

Stole the graphic from a cheap Chinese firework box.

 

[Folterknecht]

Let's do some math together, and start seeing if our BS detectors go off...

Using consumer prices (inflation of ~3%) to tackle that "problem" sure raises my bull-shito-meter, especially for companies operating mostly outside of the US or europe (production, suppliers ...).

I dont claim to know the numbers, but I doubt its a calculation u can do on a napkin or a short forum post.

 

[mastrdrver]

Editorial / Opinion. Keep it civil.

I think this in not necessarily an AMD or nVidia thing, but the problems of getting the process smaller at this point. I think because of the stalling in node shrinks because it's getting harder and harder, these type of things are going to get worse and worse. I'd be surprised if anything came after TSMC's 16nm (or what ever it is) process until at least 5 years later.

 

[lilhasselhoffer]

Using consumer prices (inflation of ~3%) to tackle that "problem" sure raises my bull-shito-meter, especially for companies operating mostly outside of the US or europe (production, suppliers ...).

I dont claim to know the numbers, but I doubt its a calculation u can do on a napkin or a short forum post.

You'll note some of the things I stated as given assumptions.


The first big ask is that inflation is a constant. Inflation varies year to year, in sections of the market, and even then we aren't accounting for the market crash of 2008. If you somehow believe that my numbers were 100% accurate, you failed to grasp that I was setting up a rough estimation. Assuming you wanted 100% accurate numbers you'd have to review governmental reporting for each year, and do the math for each individual year. That 100% accurate answer takes more than 4 times the effort of my 90% accurate answer. If you'd like to do that extended math, help yourself.

Company headquarters, and even manufacturing facilities don't matter. To the consumer the manufacturing plant could be down the street, half way around the world, or in a parallel dimension. The manufacturing facility influences only the associated materials cost for the product, as the resources used to get the goods to point of sale are lumped into the gross price. Adding in currency conversion rates is a needless complication, because they don't matter. The consumer only sees the shelf price. We are unconcerned with the price the manufacturer actually pays (and their subsequent profit margin).

Currency doesn't matter. To have to state this is stupid, but relative currency value fluctuates daily. Despite this, the cost of a card doesn't fluctuate. Fluctuating currency values are built into the selling price of cards. Even then, value fluctuation is generally insignificant. Assuming a 10% relative fluctuation in the relative value of currencies, the parent company eats a loss somewhere and a gain in another location. 2+3 = 7-2 = 900-901+6.

At this point I'm supposing that you want to factor in something else needlessly complex. How about AUD =/= USD =/= EURO =/= German Mark? The reason I chose one currency is because it makes things easy to relate. If you want to be pedantic I suggest you start calculating the relative value of Franks, Pounds, AUD, USD, Marks, and a dozen other Western European countries old currencies. Kinda seems like you're looking for a justification as to why I might be wrong then, without regards for contents of the argument.



At this point, I've justified my reasoning. It's time for you to do the same. Assuming you have no argument, I'll assume you've acquiesced to the point. If you can come back an prove that my assertion that pricing is actually in line with inflation, I'll gladly admit that I am wrong. It's your move @Folterknecht .

 

[AsRock]

Exactly my point...

8800 Ultra was $830
8800GTX was $600-$650...

... that was 9 years ago.

you can go a little further than that lol.
http://www.cnet.com/products/ati-radeon-9800-xt/#!

 

[mirakul]

Given that Win10 and Dx12 is so close, I hope that there would be a rebench on Win 10 and later a bench on Dx12 patch of Witcher3, Batman AK, Project cars.

Kind of surprise that TPU's review has been done on Win 7 for quite a long time

 

[HumanSmoke]

you can go a little further than that lol.
http://www.cnet.com/products/ati-radeon-9800-xt/#!

It has always been this way. The 7800GTX512 and it's ATI counterpart, the X1900 XTX caught flak at the time for pushing the $600-650 mark by people relatively new to tech.
For some people like myself, who got their start some time earlier, there has always been a high end. Exhibit #1 from 1998, a Quantum3D Obsidian² X-24

According to this U.S. inflation calculator, that $600 now equates to almost $900. At that was considered a bargain compared with the Obsidian Pro 100DB-4440 which was four times the price.

 

[AsRock]

Yup, my nVidia 7800 \ 7900 were around the $400 mark two cards that i wish i never owned lol.


How ever price of living a lot higher.

 

[arbiter]

think there were two big problems that lead to the disappointing Fury X results. AMD's hype and AMD's ego.

Most hype around the card was AMD fans hyping it up not AMD themselves. Most that hype was based on rumored specs of the card which shows a lot of them haven't learned a thing over last 4-5 years that specs of something don't mean its gonna be stomp anything else in its class. Specs looked like to some on paper to be a monster but turned out not to be so much in practice.

 

[johnspack]

I just want to know when nvidia will release the big daddy chip at consumer level. This titan stuff is crap. I need 970 prices to go down, arg.

 

[oldskooler]

From the greater collective community pov, most would agree, AMD didn't get it right. They rushed the product to market ( their own words ) and didn't have time for some features. I think the best review out there now would be HardOCP's review of the card. No DVI and HDMI 2.0 is a deal killer for many. I cannot use my 4k 40" Samsung that has a cheaper price and better PQ than nearly any monitor out there but I cannot use my Catleap's which were dirty cheap and have great performance. I think this card would have been great back in 2013 / 14.

 

[1c3d0g]

The rumored $850 probably and the 980Ti would have been a September launch. That way both companies would be selling hi end cards with a pretty nice margin. If things where different, that would have been the case.

Yeah.

But Nvidia is trying to throw AMD out of the mid-hi end market, before AMD comes out with Zen. They did it with 970, they did it again with 980Ti.

Yep again.

Intel iGPUs get more EUs and more cache (Kaby Lake offers 256MB cache in one model), if AMD produces a good Zen cpu core, it's APUs will get a nice boost just from that. Add HBM and things look bad for every discrete card under $150-$200.

That last part of the sentence is not so easy to say. GPU makers have an uncanny ability to double, even triple low-end GPU performance if/when integrated GPU's start performing too close for comfort. Focusing purely on Intel here, every time they figure out a way to boost iGPU performance, the established GPU makers just push the performance envelope even further. Intel is dedicating tons of die area to their GPU efforts (usually half of the die is reserved just for the GPU alone!), in most generations they manage to DOUBLE performance, but the GPU makers always come back and smash them head on, leaving them dizzy so they have to go back to the drawing board to come up with another plan. So Intel will always play catch up, a constant game of cat & mouse, unless it buys/merges with a well-respected GPU maker.

 

[newtekie1]

Most hype around the card was AMD fans hyping it up not AMD themselves. Most that hype was based on rumored specs of the card which shows a lot of them haven't learned a thing over last 4-5 years that specs of something don't mean its gonna be stomp anything else in its class. Specs looked like to some on paper to be a monster but turned out not to be so much in practice.

No, there was a crap ton of AMD hype.

Examples:

AMD Claims Fury X is faster than 980Ti at 4k. = False

AMD Claims Fury X runs at 50°C under typical load. = False

AMD Claims Fury X under load is less than 32db. = False(but close at least)

AMD Claims, in the first 5 minutes, Fury X is the worlds fastest graphics processor, world most power efficient graphics processor, and allows revolutionary form factors. = Again all False

 

[RejZoR]

not true at all, you dont remove tearing if you cap, you only remove it if you sync

people should be capping/syncing in the first place for years, i know i am, why should any hardware uselessly add more fps (excluding competitive)? or worse, why should anything at any time go to 3,000 fps in a menu!?

anyway you dont 'bring down power consumption' if many sites are testing at 4k with fps in the 40s, you only bring it down if you're originally going past 60 for sustained periods (or whatever monitor refresh you have)


proper enthusiast right here

True, you dont't eliminate it entirely, but it's a point for tearing where you have framerate below the screen refresh and above it. And that does make a difference.

 

[arbiter]

No, there was a crap ton of AMD hype.
Examples:
AMD Claims Fury X is faster than 980Ti at 4k. = False
AMD Claims Fury X runs at 50°C under typical load. = False
AMD Claims Fury X under load is less than 32db. = False(but close at least)
AMD Claims, in the first 5 minutes, Fury X is the worlds fastest graphics processor, world most power efficient graphics processor, and allows revolutionary form factors. = Again all False

Well those were said during the official anouncement, there was a TON hype by AMD fans then numbers of 4k didn't help but probably AMD felt they had to do make the numbers look good. AMD did put their foot in their mouth with those but still most hype and let down was fans end.

 

[newtekie1]

Well those were said during the official anouncement, there was a TON hype by AMD fans then numbers of 4k didn't help but probably AMD felt they had to do make the numbers look good. AMD did put their foot in their mouth with those but still most hype and let down was fans end.

Yes, but they were all said before the official launch, about a week before actually. The fan hype was really just people taking what AMD said and running with it.

 

[HTC]

No, there was a crap ton of AMD hype.

Examples:

AMD Claims Fury X is faster than 980Ti at 4k. = False

AMD Claims Fury X runs at 50°C under typical load. = False

AMD Claims Fury X under load is less than 32db. = False(but close at least)

AMD Claims, in the first 5 minutes, Fury X is the worlds fastest graphics processor, world most power efficient graphics processor, and allows revolutionary form factors. = Again all False

Unfortunately, i have to agree.

All the talk of performance over watts without giving any actual performance figures (and referring to their biggest single GPU card power guzzler too) made me suspect something like this but i really hoped they learned from "the bulldozer hype fiasco": it seems they didn't

The card is good and is a definite boost over their previous single GPU card high end, both performance as well as power usage wise but it fails quite hard because AMD hyped it WAY too much, making it seem much better then it actually is, and it's byting them in the ass, with great white shark teeth ...

EDIT

@OP: since this is an editorial / opinion, it shouldn't be in news section, IMO.

 

[Ebo]

Fore me the situation is a bit "difficult".

I see it as, AMD with Fury X, has shown that the HBM tek works, thats the way to go in the future, thats where the money is.

As far as preformance, thats a bit dissapointing which is mostly due to the hype arround Fury X. Had Nvidia launched GFX 980TI in late summer(as planned), we would all just sit and say....WOW what a card, but Nvidia pulled out a rabbit.

I know most gamers play in less resolutions, to them, its a entusiast card, its NOT made for the masses, its made for the top of the pop.
You might argue that the price is too high, I can only say get over it and find the card in your pricepoint that suits you, I couldnt care less.
The price is right and people are buying it so far, now we can only wait and see what R9 Fury on air brings to the table. That card might end up being the real bang for the buck for AMD, just like R9 290 is, great preformance for a reasonable price.

As for 4K gaming, thats just too soon, the screens just arent good enough yet, but they will be eventually. If you are in the market for a card like Fury X or GTX 980TI it would be a real shame to let them run at a 1080 screen, its simply not good enough to being able using all that power, the sweet spot is in my opinion is 2560X1440 as of now..

 

[kiddagoat]

I dunno..... seems like @newtekie1 is the sort of passive aggressive fanboy..... I mean I don't hear any constructive just constant bashing without all the flare and flame usually associated with the die hard fanboy......

Yeah AMD hyped up the card, they most certainly did.... I mean that's the PR and marketing team's job... they generate buzz... Nvidia does it, Intel does it, MSI, ASUS, on and on... all vendors do it.... Looking at the raw numbers yeah the 980Ti is ahead... but you are talking 5-10FPS at their targeted resolutions which are 1440p and above.... which in most cases turns out to be a wash because they trade back and forth.... yeah lower resolution the gap is larger but these cards aren't made for 1080p or lower.... they aren't marketed to that...

Again this is a case of people just looking for any little thing to bash a company on.... I mean does it really make you feel that much better to come here and flame?? I sometimes am amused by these comments but some of these are just disheartening both as an enthusiast and an engineer.... wtf...... It is as if people think coming up with these technologies is easy and that anyone can do it...

 

[qubit]

How about that high pitched pump noise we read in reviews ... did they fix it? Is it noticeable with your card?

No noise at all.... I have a Sapphire... Card is audible... silent. I had to look inside the case to make sure the fan was running when I turned my machine on.

I'm very glad the noise isn't a problem like in the review.

In that case, it looks like AMD shot themselves in the foot by delivering a faulty unit to TPU for review. I think they would do well to send another one so the review can be updated, since a problem like that is a dealbreaker.

Enjoy your card.

 

[Frick]

In that case, it looks like AMD shot themselves in the foot by delivering a faulty unit to TPU for review. I think they would do well to send another one so the review can be updated, since a problem like that is a dealbreaker.

Enjoy your card.

I think other reviewers had the same problems, could be a bad batch.

 

[techy1]

nice title Picture..., but needed sume update... sadly next time (after year or two) in this Picture - there will be no Red or Green fists, there will be only the Green hand full with our green Money (cuz grean team will asky any price without competition). RIP AMD :'( you were great and we all needed you (tough you sucked last years)

 

[Ebo]

#74

What rock do you live under ?

AMD will live no matter what, even Nvidia needs them otherwise there will be no development regarding GFX, then all is over. Nvidia hasent come up with 1 single breaktrough in years, they have only been on the wave AMD)ATI started. Im actually not a fanboy, but ive had AMD/ATI ever since the battle between ATI 8500 and Gforce 2.
Ive sticked with AMD ever since before catalyst drivers were envented, but im still not a fan boy, saying AMD is best.
Ive only gone that way because I know how to tweak the drivers for AMD, and I have no clue how to do that on a Nvidia card, and I really dont care. Im happy about the choices ive made and i will stick with that no matter what as long as my card deliveres and preform the way I expect it to.

 

[qubit]

It has always been this way. The 7800GTX512 and it's ATI counterpart, the X1900 XTX caught flak at the time for pushing the $600-650 mark by people relatively new to tech.
For some people like myself, who got their start some time earlier, there has always been a high end. Exhibit #1 from 1998, a Quantum3D Obsidian² X-24

According to this U.S. inflation calculator, that $600 now equates to almost $900. At that was considered a bargain compared with the Obsidian Pro 100DB-4440 which was four times the price.

Wow, that's a blast from the past! I can see that the price is eye watering, too.

That looks like a daughterboard at the back there. Do you have a clearer picture of it?