- Joined
- Apr 10, 2003
- Messages
- 23,511
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
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.
7990 first shots fired.
- Thread starter cageymaru
- Start date
Sprayingmango
[H]ard|Gawd
- Joined
- Jan 20, 2012
- Messages
- 1,259
AMD wishes it was still relevant. Micro-stutter and frametime issues need to be fixed before anyone will care. So that card has slightly more power than a single Titan? LOL!
- Joined
- Apr 12, 2003
- Messages
- 7,534
Eh... 
need to work on frame times and such.
need to work on frame times and such.What's not to get; they are comparing a thousand dollar card against another thousand dollar card. The Titan is without a doubt the best single card on the market, but it is greatly overpriced.
Won't disagree with possible frame or micro-stutter issues, but judging by the graph on hardware info it blows the Titan out of the water on Battlefield 3 so it's a little more than slightly.
- Joined
- Jul 29, 2011
- Messages
- 7,453
Titan probably has more headroom in overclocking to narrow the gap, that seems to be the downfall of the dual GPU cards. I'm not too interested in 7990 if its anything like 6990: hot, loud, and full of microstutter.
17fps increase over a single titan isn't really blowing it out of the water, especially when micro stutter is present on that 57 fps.
It's comparing it against the 690? They even say it,
"The results show that the Radeon HD 7990 is a few FPS faster than its competitor aka NVIDIA GeForce GTX 690"
TheTitan is compared in the 3D mark test, and in BF3 it shows a few different cards.
I just wonder did you complain in benchmarks where they tested a $1000 card against a $450 card?
Eh...
I have heard on other forums that AMD will be releasing their long awaited memory management driver with the release of this card, promising a reduction o frame times and stuttering generically across Dx10 and DX11 titles.
Hopefully this will be the case. It would be nice for them to sort out crossfire once and for all.
BababooeyHTJ
Supreme [H]ardness
- Joined
- Jan 21, 2009
- Messages
- 6,951
Not really, titan isn't exactly the greatest clocker. The cards used in review probably mostly boost a good bit over 1000mhz, probably close to 1050mhz. Most retail cards won't hit 1200mhz completely stable. Mine only did 1137mhz completely stable and that was under water with a tweaked bios.
I would assume that 7990 will have voltage control much like the rest of AMD's lineup. So with good cooling (water) you can be pretty certain that the 7990 will be the better clocker.
Then there is the microstutter debate. Its there, no doubt about it. I've seen it. Thankfully its fixable. For the most part this card will be pumping out a much higher framerate. So if you're using a 60hz display using a framerae cap is a very viable workaround. I would use one on a GTX690 if I were using a 60hz display too.
I didn't find microstutter to be an issue at 2560x1440@60hz thanks to a framerate cap which will even make the game smoother on a single card if you can keep the framerate that high. Once I picked up a 120hz 1440p display trying to maintain that high a cap wasn't a viable workaround. I suppose that eyefinity users might have the same problem.
The fact of the matter is you can use higher settings with a framerate cap with this card than you can get away with using with a titan as long as you're using a 60hz display.
Guess you missed it, but, they based the title of the article on the 3Dmark scores where it does get much higher than the 690. They updated the graph from hardware info at a later time.
and if the rumoured price is true, around $700 the it will be a very tempting card. But of course, only if the driver does all it is supposed to.
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
um that's a 42% increase so yes that would be blowing it out of the water for the same price or cheaper. of course the microstutter is an issue though.
I agree with all your post, I think the 7990 will be a good overclocker too. All the GCN based cards seem to have great headroom, don't see why this one won't have. I hear some people are getting great overclocks with their Titan's so not sure which of the two will be better. Guess we will know in 3 days.
Oh and just in case you missed my last post, you might not have to do the frame cap trick anymore. There is talk of a new driver that's going to fix the frame rate times and stuttering on DX10 and DX11 titles. It's supposedly going to be released around the same time as the 7990. Hope it's true.
Yes, the price point range enabling a more diverse market to access it was the right move for AMD.
BababooeyHTJ
Supreme [H]ardness
- Joined
- Jan 21, 2009
- Messages
- 6,951
That sort of difference might be the difference between 4x msaa and 2x msaa or even no aa and still being able to maintain a cap.
Yeah, it would be an amazing buy at that price. I would be surprised if that were true though. Just based on current prices.
Somebody on another forum saw it for sale at that $700 price but it was taken down really quickly. Of course they could be lying
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
it depends. if they don't have a lot of product then yeah $899 sounds about right. they will want to undercut and be faster than Nvidia based on the fact that they clearly are behind when it comes to smooth dual gpu performance. if they have a lot of product then $799 and possibly $699 may not be unrealistic. I am guessing $799 myself.
Titan overclocking blows. Stop making up random stuff
That's not a really good way to look at it; a better way is to recognize that the 17 frames represents an increase of 42.5% which at least in my book is dominating. As far as the possible micro stutter, run Radeon Pro and see what you have at that point.
But with the frame time issues present and currently known, that isn't a realistic representation of that card's FPS. Thus I would honestly say it's probably closer to being around the same fps as the 690, ASSUMING this dual gpu card has high enough frame latency, of course.
A good quote from the Bioshock Infinite benchmark review,
Source
5150Joker
Supreme [H]ardness
- Joined
- Aug 1, 2005
- Messages
- 4,568
I think the one making up stuff is you. A stock titan is 876 MHz, I can get both of mine to do 1100+ on stock voltage. With a tiny bump, I can get them to do 1176 MHz on both. That's +300 MHz core overclock or 34% increase on air. Since when does that constitute blowing?
And as far as boost, I have a vbios that does away with GPU boost 2.0 and the clocks are 100% stable at 1176+ MHz. Here's an example on my old Asus Titan that wasn't even a good overclocker: http://forum.techinferno.com/nvidia-video-cards/3343-gtx-titan-overclocking-bios-mods.html
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
nice try but do you really think we are that dumb? lol. comparing your max boost oc to an advertised boost clock? really? a stock Titan typically goes to around 1000 mhz under boost and that is the performance we see in reviews. so your actual oc is about 15-17% over that.
5150Joker
Supreme [H]ardness
- Joined
- Aug 1, 2005
- Messages
- 4,568
Since when was anyone talking about the programmed boost? He said Titan OC sucks and I addressed that. FYI 1176 MHz is not my max OC, I can comfortably hit 1200. Others with water are doing 1230+. As far as my OC over stock boost, "typically" isn't a way to figure a percentage, you have to use the base clock since boost is variable from one card to the other.
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
come on now. the reviews show performance with actual stock boost being right around 1000mhz so our "oc" is not anywhere near the percentage you claim. not even but about half of it.
5150Joker
Supreme [H]ardness
- Joined
- Aug 1, 2005
- Messages
- 4,568
You can't base a percent OC off a boost clock, that's nonsense. You have to always go by the base clock. Even the stock boost is an OC itself so you don't calculate a further OC off an OC.
Since when does anyone NOT account for boost? You affirmed what I "made up" your Titan OC sucks. Heck youre likely throttling back from that already crappy OC. Don't worry, my 680 OC like crap too. The only reason not to account for boost is if you want to pretend you got a good OC. In reality you got peanuts over a stock Titan... That's what you measure against, stock, not "base"
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
congratulations on making a fool of yourself. get through your head that reviews show the performance of what the card is actually boosting too which is about 1000 mhz. your damn overclock to 1176 is NOT a 34% oc.You can't base a percent OC off a boost clock, that's nonsense. You have to always go by the base clock. Even the stock boost is an OC itself so you don't calculate a further OC off an OC.
5150Joker
Supreme [H]ardness
- Joined
- Aug 1, 2005
- Messages
- 4,568
No throttling at all, modded vbios that removes it completely.
congratulations on making a fool of yourself. get through your head that reviews show the performance of what the card is actually boosting too which is about 1000 mhz. your damn overclock to 1176 is NOT a 34% oc.
What a review shows or doesn't is irrelevant. The fact is the stock clock is 876 mhz and anything above that (aka boost) is an OC or rather "auto clock" upwards from base up to the stock bios TDP. The statement was, "Titan OC is crap" not "Titan OC is crap over boost on a stock vbios". You guys are shifting your argument. If 100% of Titans reached 1000 MHz boost under every system setup out there, then we could establish 1000 MHz as a baseline, not just from a handful of reviews out there.
But lets say we do go with 1000 MHz as a reference clock, hitting 175+ MHz over that is hardly "crap". You guys have very lofty expectations.
P.S. What do you call it when GPU Boost is disabled as in my case with the custom vbios? Base clock is 876 MHz and there is no boost at all + power limits at 350/380 Watts.
Last edited:
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
lol keep on looking silly. the fact is the cards are boosting to around 1000 so the advertised 876 does NOT mean a damn thing. your ACTUAL oc is only above what the card was boosting too not the advertised speed.
EDIT: and again the POINT is that the performance you see in reviews is with cards hitting around 1000 mhz so your actual "oc" to 1176 is about 17% or so.
Last edited:
Well you're one up on most Titan owners. You got a crappy OC with no throttling while most have to contend with throttling.
BababooeyHTJ
Supreme [H]ardness
- Joined
- Jan 21, 2009
- Messages
- 6,951
That is a very small percentage of Titan owners and I bet that your oc isn't completely stable. My card could loop Heaven for endless hours at 1200+ but Far Cry 3 would make it crash at anything over 1137mhz. Some cards do a little more, some do less.
I don't believe every person that I see bragging on the forums.
Yup, and I bet that you could find a few games to crash his oc.
Back to your cave troll.
- Joined
- Feb 15, 2011
- Messages
- 2,877
I agree with you, nVidia fanboys out in full force already, but be careful on calling out specific trolls.. you might get an infraction.
That's_Corporate
[H]ard|Gawd
- Joined
- Sep 30, 2006
- Messages
- 1,738
That's only because AMD set the bar high in terms of overclockability (if that's a word hehe).
you just like to argue with everyone.
Since not all titans boost the same you have to go off there base clock. that's the only baseline from titan to titan. that's apples to apples
your talking apples to dog shit.