Grooveriding
Diamond Member
- Dec 25, 2008
- 9,147
- 1,330
- 126
Keysplayr
Elite Member
- Jan 16, 2003
- 21,219
- 56
- 91
Ahem...... :: clears throat :: tap tap tap - Is this thing on?
The only thing that isn't good about this GPU is it's price. It's way up there.
After that, you'll be hard pressed to REALLY find anything lackluster about it.
I mean, you can make stuff up, but that's pretty transparent. The only issue
I'd agree with is the price and the performance one gets for that price. That's it.
Last edited:
DooKey
Golden Member
- Nov 9, 2005
- 1,811
- 458
- 136
I agree. There's no disputing that the performance is outstanding compared to what else is on the market, but the price is too high for that performance in my opinion. Other than that it's a nice piece of engineering and a complete step above all the other single GPU cards on the market. Anyone knocking this card for anything other than price has an agenda.
Grooveriding
Diamond Member
- Dec 25, 2008
- 9,147
- 1,330
- 126
It's a nice card, the best flagship improvement since 8800GTX. It doesn't run all that cool as purported, but I could care less about the real world noise/heat as I watercool.
Just don't understand the price, it's not good enough. Even $900 when that was rumoured seemed like too much, but then the rumour was 50% as fast as a 680. Now it's a $1000 and a little less than 40% in reality... The price is ridiculous.
Just don't understand the price, it's not good enough. Even $900 when that was rumoured seemed like too much, but then the rumour was 50% as fast as a 680. Now it's a $1000 and a little less than 40% in reality... The price is ridiculous.
wand3r3r
Diamond Member
- May 16, 2008
- 3,180
- 0
- 0
Yes, it's a good card. Nothing to refute there.
It's not a god-like card like they are trying to price it though.
At the normal price (under $600) it'd fit right in relative to the current situation, and it'd actually be a pretty good value. I just hope the market reacts and the price drops to sane levels quickly because it'd be a great card to have.
On topic, these reviews are appearing to confirm the leaks in the end of the original thread.
in case you forgot: standard deviation is a measure of dispersion from average.
smaller deviation -> distribution is closer to avg.
Last edited:
40% faster than the curent fastest single GPU (7970 @ 1050MHz) = great performance.
180% price premium for ~40% performance = ripoff
180% price premium for ~40% performance = ripoff
Grooveriding
Diamond Member
- Dec 25, 2008
- 9,147
- 1,330
- 126
Keysplayr
Elite Member
- Jan 16, 2003
- 21,219
- 56
- 91
Not even 40%.. 29% over 7970GE and 36% over 680...
LOL you pick the lowest numbers to show.
Maybe you could take a look at all the other graphs being shown.
Grooveriding
Diamond Member
- Dec 25, 2008
- 9,147
- 1,330
- 126
No need to get sensitive, I linked whatever I found earlier showing better individual scores in some cases.
Note the 'Average Improvement Over' HWC standard conclusion they use in all their reviews. Taking all their results and showing you the average improvement from one card to another across their benchmark suite. So that includes the 40% improvements, as well as the 15% ones...
Looking forward to seeing computerbase.de's overall performance conclusions when reviews go live later as well :thumbsup:
Lonbjerg
Diamond Member
- Dec 6, 2009
- 4,419
- 0
- 0
I think it will depend heavily on the game. Titan and 7970GE have the same memory bandwidth. If that becomes a bottleneck, they could be quite close together.
blastingcap
Diamond Member
- Sep 16, 2010
- 6,654
- 5
- 76
Not even 40%.. 29% over 7970GE and 36% over 680...
I guess it depends on resolution because 29/36 are for the 25x16 rez. It's 25/45 for Eyefinity rez, though how much of that is the 680 running into a VRAM wall I don't know.
Grooveriding
Diamond Member
- Dec 25, 2008
- 9,147
- 1,330
- 126
Yeah I think memory bandwidth situations are where Titan will look its best. That is mostly going to be older games that are really demanding there, more modern games like Far Cry 3, Crysis 3 & Battlefield 3 are not that memory bandwidth hungry.
Older games like Metro 2033 and Crysis 1/Warhead I expect to show the best results on Titan, where GK104 based cards generally fall flat due to low memory bandwidth.
Keysplayr
Elite Member
- Jan 16, 2003
- 21,219
- 56
- 91
No need to get sensitive, I linked whatever I found earlier showing better individual scores in some cases.
Note the 'Average Improvement Over' HWC standard conclusion they use in all their reviews. Taking all their results and showing you the average improvement from one card to another across their benchmark suite. So that includes the 40% improvements, as well as the 15% ones...
Looking forward to seeing computerbase.de's overall performance conclusions when reviews go live later as well :thumbsup:
If you know that much, then you also know the spread of the percentage spectrum doesnt max at 40% as you allude to. Is 40% The highest you've seen in previous graphs? I'm looking again right now.
This, and more @ high-resolutions. But what I really want to see is compute performance, how much it improves over GK104.
Performance wise is nothing AMD cant reach with a refresh, and if they really go for it, it will be funny to see how the price of this goes down to $ 500/600 levels. :awe:
Grooveriding
Diamond Member
- Dec 25, 2008
- 9,147
- 1,330
- 126
I have no idea, it was a general remark to help you understand that graph indicates the card's overall performance improvement across their benchmark suite.
If you're saying you want to see the best possible case game to cherry pick, then I'm sure you'll be able to find outliers where it exceeds the average of 36%>GTX680 29%>7970GE. Just like you'll find games where its improvement is even lower than the average. Hence, it's an average.
Keysplayr
Elite Member
- Jan 16, 2003
- 21,219
- 56
- 91
Tempered81's posted graph shows 6 games where Titan has greater than 40% advantage even a few over50% and one even over60%. As far as your lower end of 15%, only two titles were that low at 15 & 16%.
I'm not saying I want to see the best possible case. What I want to see is you and others keeping it real.
That is all. Also you saying I'm being sensitive is ridiculous. A diversion. It just isnt so.
I'm not saying I want to see the best possible case. What I want to see is you and others keeping it real.
That is all. Also you saying I'm being sensitive is ridiculous. A diversion. It just isnt so.
Last edited:
hyrule4927
Senior member
- Feb 9, 2012
- 359
- 1
- 76
Yes, how dare we discuss the average improvement. Clearly only cherry picked benchmarks accurately represent this card.
Keysplayr
Elite Member
- Jan 16, 2003
- 21,219
- 56
- 91
Yes, how dare we discuss the average improvement. Clearly only cherry picked benchmarks accurately represent this card.
Riiiight..
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 24K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter