LOL.
I thought I was bad. I may just cut my order down to one of these just to play with in the meantime though
Naa dont do that, just send me 1 and heck just because im such a cool dude i'll even cover the postage.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
AMD Radeon R9 290X with Hawaii GPU pictured, has 512-bit 4GB Memory
- Thread starter LtMatt
- Start date
Naa dont do that, just send me 1 and heck just because im such a cool dude i'll even cover the postage.
those charts seem to say buy 2 7950's and save some money
Thats exactly what I'm thinking
Let's play with some scenarios in dual card first, so these are the scores from our firestrike thread.....
11503. GFX Score 14326. 670. Air cooled. Physics Score 13270. CPU 2700K @ 5.2 Air cooled - DiaVolaAce - Link
15700. GFX Score 19332. 7970. Water cooled. Physics Score 17400. CPU 3930K @ 5.0 Water cooled - Besty - Link
Based on this information, you can speculate that with a 37.5% performance uplift over a 7970 a dual 290X setup under water with 1.35v-1.4v will produce an additional '7250' graphics score over my cards, so lets round that down to 7000.
7000+19332 = 26,500 'ish (rounding down again).
A score in that ball-park a heavily over-clocked 290X would be 83% faster than a the example 670 SLI setup with OC and trading blows with the titan.
Dial the overclock back on the 290x and you are sill way beyond a 20% uplift over the 670SLI, even accounting for differences in CPU speed on the example rigs.
Using a single card, and assuming the 290x is overclocked, you are looking at a 40-45% uplift over the 670SLI ?
Assuming the 290x is at stock well, its going to be interesting to find out, you are probably correct but its going to be by small percentages and the non-reference cards may get it to 21% at some point but I think it will be too hot on air to do so.
Last edited:
Soldato
- Joined
- 6 Feb 2010
- Posts
- 14,583
Not computer museum, but I can imagine something like a shoe-wears store, but with graphic cards instead of shoes hanging on the walls, and boxes and boxes of graphic card on the shelfsKaap could probably start a high-end computer museum!
Soldato
- Joined
- 22 Jul 2012
- Posts
- 16,185
- Location
- London
Most of them are in pairs too...
Kaaptown?
Kaaptown?
More like wishful thinking
Can I ask why with a 280x and not a 290x?
Thanks for the info
Soldato
- Joined
- 6 Feb 2010
- Posts
- 14,583
Because 280x is a 7970 rebadge, so it is the same GPU architecture; the 290x on the other hand is based on the new Hawaii architecture, so it shouldn't be able to crossfire with the 7970 or 280x.
The 280x is a pure rebrand of the 7970 meaning it fits right in crossfire. The 290x on the other hand is not a rebrand and the architecture is different. As fare as ive heard and seen and what common logic dictates GCN 1.0(7000 series cards) cannot crossfire with GCN 2.0 (290 and 290x).
Caporegime
- Joined
- 30 Jul 2013
- Posts
- 30,366
Plus the 290/290x doesn't have a dedicated crossfire connector anyway.
It's now all done over PCI Express.
It's now all done over PCI Express.
Besty, you have picked the fastest 7970 and picked a random 670 score...Hardly a fair playing field, so I decided to run some numbers...
670 SLI score of 15506
7970 CF score of 19332
That makes your cards 28.5% faster than the 670 SLI. Now if we take my heavily overclocked Titan and see how that gets on against both those systems, I got 13825 GFX score.
So we have 670 scoring 15506
and we have a Titan scoring 13825
That is 12% slower than those 670's in SLI add on 20% to the Titan and add the original 12% (32% in total) and the 290X needs to score 18249 in GFX score for you to be correct....Not a chance in hell
670 SLI score of 15506
7970 CF score of 19332
That makes your cards 28.5% faster than the 670 SLI. Now if we take my heavily overclocked Titan and see how that gets on against both those systems, I got 13825 GFX score.
So we have 670 scoring 15506
and we have a Titan scoring 13825
That is 12% slower than those 670's in SLI add on 20% to the Titan and add the original 12% (32% in total) and the 290X needs to score 18249 in GFX score for you to be correct....Not a chance in hell
Soldato
- Joined
- 2 Jan 2012
- Posts
- 12,453
- Location
- UK.
I'd usually agree with this for most yank reviewers but this guy never beats around the bush, most of his views are exactly like just you and me the consumer.
1. He says it isn't a TITAN killer which it isn't as it needs a massive percentage lead to be classed as a killer.
2. He says mantle needs to be successful to create competition.
I think most of us here can agree with those 2 main points he made.
See Suarez big-upping the vid, imagine my surprise when I give it a watch and he's slating AMD.
So we have 670 scoring 15506
and we have a Titan scoring 13825
That is 12% slower than those 670's in SLI add on 20% to the Titan and add the original 12% (32% in total) and the 290X needs to score 18249 in GFX score for you to be correct....Not a chance in hell
a 290x does 15k, sli however dosnt scale as well in many games as it does in that benchmark. so overall a 290x OC is the better option.
That is 12% slower than those 670's in SLI add on 20% to the Titan and add the original 12% (32% in total) and the 290X needs to score 18249 in GFX score for you to be correct....Not a chance in hell
haha, I don't know, 9600-9800 was the rumour for the 290x in Uber mode, double that for CF and its well, close. Too close for me to call so I say the man is right.
Not concerned with the performance, I am not concerned with the heat output. 6+8 though and not 8+8 so its gonna be fine right?
I am surprised nobody has jumped in with quad 780 classifieds yet, they seem to be producing strong performance.
Last edited: