the more i read, research, and network the more this SLI stuff seems like BAD idea. I think im gonna go with one 6800 ultra(BFG prolly unless I can get several good responses on why eVGA would be a better brand) and upgrad to one of the next gen cards when they come out. Gonna have to start putting some money away if im gonna have enough to get one when their available.
-
We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.
X800XL SLI vs One vs 6600GT SLI vs One Any bench tests?
Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Crescent13
Diamond Member
Ati DOES NOT have SLI. NVIDIA ONLY!
deadseasquirrel
Golden Member
dood, you've been here about a year now. surely you know that *any* discussion turns into an amd-intel\nV-ati\lcd-crt flame war.
but, hey, at least you and i have kept out of it.
dheffer's a fanboi!
/runs away
rise
Diamond Member
if you want to pay the extra for an ultra as opposed to a gt that will probably clock the same for less money, then go ahead. if you want to pay extra for a bfg for a warranty you admit you'll never reach, go for it. if you think sli is a bad idea, well then don't get it.
just hit the buy button already.
BenSkywalker
Diamond Member
The quote I replied to-
I responded by saying that was true, there just wasn't any great LCDs. If you disagree with that POV then tell us what is a great LCD for gaming?
otispunkmeyer
Lifer
HOLY SH!T:Q
lavaheadache
Diamond Member
I have a samsung 710n, and I think this this is great for gaming. 12 ms response and 600:1 contrast. I see no ghosting and every thing is bright and vivid
What is "good"? It's entirely subjective. One monitor might be nice for one person, but horrible for another. Let's get over this.
What's important is this kid spent all this money that way he can SLI in the future, even though in this very post, he talks about what a bad idea it is.
Glad we stayed on focus!
Its not that SLI is a bad idea period. It just seems to be a bad idea right now. With next gen cards that are around the corner and appear to be twice to thrice current cards, it would be silly to buy two currnent when next will be twice to thrice for half the price of two current cards. I am also very willing to dish out for top of the line and best of the brand. As for monitors, no crt will fit on my dest, and no 1,000 dollar component of any kink will fit in my budget.
TechnoButt
Diamond Member
2 Points for anyone late reading this thread like me:
1) New games rarely stay above 60fps in all scenarios, especially faster action games with complicated character models (ie, unreal). They *average* high frame rates for benchmarks, but when the action really hits - which is when you really need the graphics horsepower - framerates can drop well below 60 (even on the best SLI rigs in SLI capable games). The argument that LCD users don't need powerful graphics because of the lower refresh rates is simply wrong. On the topic of LCD panels, response times <20ms are useless paper specifications that do not represent actual performance. I'm a pretty serious gamer (participating in competitive clans) and I switched late last year from a Sony 19" FD Trinitron to a Samsung 193P. I'll put my panel up against anything on the market for gaming, side by side. So far I haven't seen anything that outperforms it in the 19-20" range in any way (contrast, brightness, viewing angle, color reproduction, and yes even pixel response in actual gaming/video situations). My lame response time on paper is 20ms but I do have full 8-bit color.. but I assure you there are many 16ms and 12ms panels advertised on the market that present considerably more ghosting, streaking, and so forth in real application. I haven't compared to an 8ms panel, but I expect the results are the same (they are based on the same 6-bit panel technology that allows the 12 and 16 specs).
2) SLI did not save 3DFX, innovation from Nvidia in the form of newer technology allowed Nvidia to bankrupt 3DFX (which gave them ownership). They didn't break out this technology until they couldn't beat ATI in a single processor. For people who think it's cost effective in any way, consider how fast video cards lose value (more than 50% depreciation per year). Do you really want to have TWO components depreciating that rapidly?
If you can afford to buy the best every 4 months, then yes SLI is the current king. If you're one of the people that buys once per year, or even worse.. as the OP suggests, upgrading nothing but hard drives for 3 years... the SLI is a bad choice. Yes, it will rock for the next 12 months. Then it will be inferior to a single card solution in the second year. In the third year, it will be so far behind the curve that it's almost ridiculous that your actual cost is $1000 for the setup over 3 years.
A better solution for most people will be to buy an X800XL at <$300 now. Buy the same price range next year (at the beginning of the year). And buy a third in the third year. You get more than acceptable performance that matches up with your monitor, etc now, in all but the most taxing moments (during which even the SLI setup slows down), and you maintain that level of performance for all three years and end up spending the same amount of money (ignoring time value of money). I've been playing the 'top video card game' since the days of Voodoo SLI.. and it's a very expensive hobby.
Lately, the games pretty much blow in my opinion, so this year I'm running an X800XL until someone comes up with a game that keeps my interest for more than 30 days (please dear God leat Unreal 2006 be as much fun as Unreal Tournament was for 3 years).
And there's my $.02
1) New games rarely stay above 60fps in all scenarios, especially faster action games with complicated character models (ie, unreal). They *average* high frame rates for benchmarks, but when the action really hits - which is when you really need the graphics horsepower - framerates can drop well below 60 (even on the best SLI rigs in SLI capable games). The argument that LCD users don't need powerful graphics because of the lower refresh rates is simply wrong. On the topic of LCD panels, response times <20ms are useless paper specifications that do not represent actual performance. I'm a pretty serious gamer (participating in competitive clans) and I switched late last year from a Sony 19" FD Trinitron to a Samsung 193P. I'll put my panel up against anything on the market for gaming, side by side. So far I haven't seen anything that outperforms it in the 19-20" range in any way (contrast, brightness, viewing angle, color reproduction, and yes even pixel response in actual gaming/video situations). My lame response time on paper is 20ms but I do have full 8-bit color.. but I assure you there are many 16ms and 12ms panels advertised on the market that present considerably more ghosting, streaking, and so forth in real application. I haven't compared to an 8ms panel, but I expect the results are the same (they are based on the same 6-bit panel technology that allows the 12 and 16 specs).
2) SLI did not save 3DFX, innovation from Nvidia in the form of newer technology allowed Nvidia to bankrupt 3DFX (which gave them ownership). They didn't break out this technology until they couldn't beat ATI in a single processor. For people who think it's cost effective in any way, consider how fast video cards lose value (more than 50% depreciation per year). Do you really want to have TWO components depreciating that rapidly?
If you can afford to buy the best every 4 months, then yes SLI is the current king. If you're one of the people that buys once per year, or even worse.. as the OP suggests, upgrading nothing but hard drives for 3 years... the SLI is a bad choice. Yes, it will rock for the next 12 months. Then it will be inferior to a single card solution in the second year. In the third year, it will be so far behind the curve that it's almost ridiculous that your actual cost is $1000 for the setup over 3 years.
A better solution for most people will be to buy an X800XL at <$300 now. Buy the same price range next year (at the beginning of the year). And buy a third in the third year. You get more than acceptable performance that matches up with your monitor, etc now, in all but the most taxing moments (during which even the SLI setup slows down), and you maintain that level of performance for all three years and end up spending the same amount of money (ignoring time value of money). I've been playing the 'top video card game' since the days of Voodoo SLI.. and it's a very expensive hobby.
Lately, the games pretty much blow in my opinion, so this year I'm running an X800XL until someone comes up with a game that keeps my interest for more than 30 days (please dear God leat Unreal 2006 be as much fun as Unreal Tournament was for 3 years).
And there's my $.02
A. What's acceptable to you and I may be two different things. Like I said, you can always buy slower cards for less money.
B. If you think you can compare SLI slowdowns to X800XL slowdowns, you've obviously never used SLI.
Not like some others. Try fishing. <thinks of boat payment, storage/ insurance cost, $50 a pop to fill gas tank, $200 fishing rods and $5.00 baits, 10mpg pulling fishing boat>
Try golf. Try big game hunting. Try skiing. Try water skiing. Try travel.
Video cards are a pretty cheap hobby compared to a lot of them.
Most of this just seems like you're trying to convince others you did the right thing?
TechnoButt
Diamond Member
I think a better benchmark is 'slowest frame rate over any 1 second throughout a longer demo', not average frame rate. I have no evidence to support my suspicions, but I'd be willing to wager there is less than a 25% difference between the slowest frame rate in taxing moments between X800XL and 6800Ultra SLI (all other hardware being best available and identical). Now, when you're talking about 16fps vs 20fps actual performance during those taxing moments, there is a big difference (even if my 25% guess is accurate), but is that difference worth $700? That's a personal decision. I'd really like to see some review sites push this perspective (testing worst case scenarios instead of average performance in a 2 minute demo over a variety of conditions). Obviously, anything far above 60-75fps is a waste with human anatomy, so what people are really after is the best of worst case scenarios. On this topic, I'd like to see Nvidia or ATI (or hell, even Matrox) develop a graphics technology that reduces polygon count on the fly. When the action is the most fast and furious, the reduction in polygon count and texture depth would have MUCH less impact on the viewer (at some point increased details are diminishing returns on fast action).
You're right, I cannot say for sure how SLI will work in the future. Estimating from the past 6 years, next years graphics card will be roughly twice as fast as this years and the following year will be again roughly twice as fast. GPU development is occuring at a much faster pace than CPU development (although the rates of both are converging as they reach the limits of silicon). Based on the history of 3D Graphics hardware and game technology, I think it's a pretty safe bet to say that the 6800Ultra SLI rig purchased today will be a joke compared to the best single card solution in the $300 price range in 2007-2008 (three years).
You're right, I cannot say for sure how SLI will work in the future. Estimating from the past 6 years, next years graphics card will be roughly twice as fast as this years and the following year will be again roughly twice as fast. GPU development is occuring at a much faster pace than CPU development (although the rates of both are converging as they reach the limits of silicon). Based on the history of 3D Graphics hardware and game technology, I think it's a pretty safe bet to say that the 6800Ultra SLI rig purchased today will be a joke compared to the best single card solution in the $300 price range in 2007-2008 (three years).
I run sli and payed less than a 6800 ultra would cost. 2 6600gt's clocked at 600mhz core and 1.15ghz mem that will blow the doors off 1 6800 ultra. Ended up spending right round 400 for both cards. Dont get me wrong, two 6800's would have been far more interesting but factoring the cost of two and then the rate of tech advances in the vid card race and it seems to me that 1000$ on two cards that will be outdated in a year would be completely retarded. I spent 400 total, a 100 less than 1 6800 ultra. In a year I wont give a rats a#$! about how much money I spent on them and wont mind giving them away and buying the fastest card or cardz.
Chode Messiah
Golden Member
you should have got 2 6800nu and unlocked the pipes
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: 25K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 24K
-
-
