If I remember correctly, Crysis is still a limiting factor for the entire system, but not "new" enough for modern graphicscards.
Other way roung, the full potential of a graphicscard cannot be seen when the system is slowing it down.
Still I agree, until the question "But can it play Crysis?" becomes a standart "Yes", it should be included in the grapgicscards tests
Originally Posted by Spreadie Antony, can you explain why you still use Crysis when testing motherboards, yet refuse to when testing graphics cards?
Crysis is a resource-hog and stresses the entire PC, this makes it excellent for jusding how much speed you're adding to a PC by overclocking its CPU, memory etc. However, this characteristic makes it a poor tool to gauge how much faster a graphics card is than another one.
it's also the case that Crysis is freakishly hard for a graphics card to run as it uses non-standard techniques. It's therefore not indicative of a graphics card's performance in any other game. Most of the games we test with share graphics engines with many other games, and they all use rendering techniques that are more standard in games development and creation. Our graphics card test games are therefore much more indicative of the level of performance each graphics card offers.
The standard response has been that "Crysis is a three year old game and graphics card tests should reflect performance in modern day games".
I agree that Crysis is old hat, and have no particular desire to see it return as a benchmark - I was never a huge fan of it anyway - it was a poorly coded and mediocre, albeit pretty, game at best.
The point I was trying to make was that given the aforementioned standard response your team used to trot out, don't you consider it unrealistic to continue using it for benching other components?
Originally Posted by phuzz Surprisingly high scores given that the overall tone of the review seemed to be "We Do Not Recommend".
That's what I thought.
Quote:
Originally Posted by Spreadie That's the first time I've seen that explanation.
The standard response has been that "Crysis is a three year old game and graphics card tests should reflect performance in modern day games".
I agree that Crysis is old hat, and have no particular desire to see it return as a benchmark - I was never a huge fan of it anyway - it was a poorly coded and mediocre, albeit pretty, game at best.
The point I was trying to make was that given the aforementioned standard response your team used to trot out, don't you consider it unrealistic to continue using it for benching other components?
Why? They're using it to stress test the whole system. While it doesn't properly reflect graphics card upgrades due to the reasons mentioned above it still stresses the CPU, memory & busses.
Originally Posted by perplekks45 Why? They're using it to stress test the whole system. While it doesn't properly reflect graphics card upgrades due to the reasons mentioned above it still stresses the CPU, memory & busses.
Yeah, thanks for that, but if I wanted a nigh-on verbatim repeat of what Clive said, I'd have just read his post again.
Regardless of the lack of SLi or preference for Crossfire. My question is why is Crossfire or SLi valid for a motherboard review, when 9 times out of 10 it's recommended against multi card setups when it comes to Graphics card reviews.
I completely understand that GPU are scaling up much better now in either format but the base recommendation is still get the best single card you can.
If that is the case, why is this board disadvantaged other than the high price.
Originally Posted by Snips Regardless of the lack of SLi or preference for Crossfire. My question is why is Crossfire or SLi valid for a motherboard review, when 9 times out of 10 it's recommended against multi card setups when it comes to Graphics card reviews.
I completely understand that GPU are scaling up much better now in either format but the base recommendation is still get the best single card you can.
If that is the case, why is this board disadvantaged other than the high price.
Well if you had the choice of going for a board that offered sli vs one that didn't for the same price, wouldn't you go for the one that did? Purely as it was offering more for your money. Now in this case, it is so much more worse as it is much more expensive.
Given there is little performance difference between the Pure Black and say the Sabertooth, to differentiate, consumers are going to look towards features offered and price. Both points where the Pure Black is severly handicapped.
Regardless of weather people use it or not, if it's a feature your compitition offer, to remain competitive, you must offer it also.
Seriously, this board is lacking on standard features, it's got poor BIOS support, it lacks additional bells and whistles you get on high-end boards, it doesn't seem to overclock too well, and it is very expensive given the competition, yet you give it an 8/10? Something tells me someone doesn't want to piss off Sapphire too much. Got any ad campaigns going right now, guys?
Originally Posted by Snips Regardless of the lack of SLi or preference for Crossfire. My question is why is Crossfire or SLi valid for a motherboard review, when 9 times out of 10 it's recommended against multi card setups when it comes to Graphics card reviews.
I completely understand that GPU are scaling up much better now in either format but the base recommendation is still get the best single card you can.
If that is the case, why is this board disadvantaged other than the high price.
Well if you had the choice of going for a board that offered sli vs one that didn't for the same price, wouldn't you go for the one that did? Purely as it was offering more for your money. Now in this case, it is so much more worse as it is much more expensive.
Given there is little performance difference between the Pure Black and say the Sabertooth, to differentiate, consumers are going to look towards features offered and price. Both points where the Pure Black is severly handicapped.
Regardless of weather people use it or not, if it's a feature your compitition offer, to remain competitive, you must offer it also.
I completely agree with you but my question was why do they recommend to not use a multi card configuration when choosing a GPU but then make it an important factor in choosing a motherboard.
Originally Posted by Snips Regardless of the lack of SLi or preference for Crossfire. My question is why is Crossfire or SLi valid for a motherboard review, when 9 times out of 10 it's recommended against multi card setups when it comes to Graphics card reviews.
I completely understand that GPU are scaling up much better now in either format but the base recommendation is still get the best single card you can.
If that is the case, why is this board disadvantaged other than the high price.
Well if you had the choice of going for a board that offered sli vs one that didn't for the same price, wouldn't you go for the one that did? Purely as it was offering more for your money. Now in this case, it is so much more worse as it is much more expensive.
Given there is little performance difference between the Pure Black and say the Sabertooth, to differentiate, consumers are going to look towards features offered and price. Both points where the Pure Black is severly handicapped.
Regardless of weather people use it or not, if it's a feature your compitition offer, to remain competitive, you must offer it also.
I completely agree with you but my question was why do they recommend to not use a multi card configuration when choosing a GPU but then make it an important factor in choosing a motherboard.
Because if the only differentiator between a two motherboards is SLI support, you might as well choose the SLI one, right?
Also, if you have an expensive motherboard you're more likely to be doing things where SLI is an advantage. Bit-Tech tends to say that SLi is a bad idea, but that's mainly (from what I gather) because the context is usually 'Should I get 2x HD5850s or 1x GTX580?'. Once you hit the top of the line, multi-gpu becomes the only way to improve, and if you want to do something like Surround, you would be wanting multi-gpu to max out 3 high-res monitors.
Originally Posted by Krayzie_B.o.n.e. yeh another overpriced under achieving board using a soon to be dead socket.
For a first attempt at a motherboard, it's pretty good.
In time, hopefully they'll produce something special - IIRC the team that's made these motherboards is the team that brought us most of the recent EVGA motherboards.
@ Bakes, if you read Instagib answer which was pretty much what you were saying. We know why you would need a mutli card compatible motherboard. GPU scaling has gone through the roof in the recent good showings from both Nvidia and AMD but.....
"I completely agree with (both of) you but my question was why do they recommend to not use a multi card configuration when choosing a GPU but then make it an important factor in choosing a motherboard."
Comments 1 to 21 of 21
ReplyOther way roung, the full potential of a graphicscard cannot be seen when the system is slowing it down.
Still I agree, until the question "But can it play Crysis?" becomes a standart "Yes", it should be included in the grapgicscards tests
Crysis is a resource-hog and stresses the entire PC, this makes it excellent for jusding how much speed you're adding to a PC by overclocking its CPU, memory etc. However, this characteristic makes it a poor tool to gauge how much faster a graphics card is than another one.
it's also the case that Crysis is freakishly hard for a graphics card to run as it uses non-standard techniques. It's therefore not indicative of a graphics card's performance in any other game. Most of the games we test with share graphics engines with many other games, and they all use rendering techniques that are more standard in games development and creation. Our graphics card test games are therefore much more indicative of the level of performance each graphics card offers.
Hope that answers your question! :)
The standard response has been that "Crysis is a three year old game and graphics card tests should reflect performance in modern day games".
I agree that Crysis is old hat, and have no particular desire to see it return as a benchmark - I was never a huge fan of it anyway - it was a poorly coded and mediocre, albeit pretty, game at best.
The point I was trying to make was that given the aforementioned standard response your team used to trot out, don't you consider it unrealistic to continue using it for benching other components?
Of course it' not my real name - how cruel do you think my parents are law99? :D
And we don't really care about Metro 2033, it's a dull game and hardly anyone actually plays it...
I completely understand that GPU are scaling up much better now in either format but the base recommendation is still get the best single card you can.
If that is the case, why is this board disadvantaged other than the high price.
Well if you had the choice of going for a board that offered sli vs one that didn't for the same price, wouldn't you go for the one that did? Purely as it was offering more for your money. Now in this case, it is so much more worse as it is much more expensive.
Given there is little performance difference between the Pure Black and say the Sabertooth, to differentiate, consumers are going to look towards features offered and price. Both points where the Pure Black is severly handicapped.
Regardless of weather people use it or not, if it's a feature your compitition offer, to remain competitive, you must offer it also.
I completely agree with you but my question was why do they recommend to not use a multi card configuration when choosing a GPU but then make it an important factor in choosing a motherboard.
Touche Mr Journalism.
Because if the only differentiator between a two motherboards is SLI support, you might as well choose the SLI one, right?
Also, if you have an expensive motherboard you're more likely to be doing things where SLI is an advantage. Bit-Tech tends to say that SLi is a bad idea, but that's mainly (from what I gather) because the context is usually 'Should I get 2x HD5850s or 1x GTX580?'. Once you hit the top of the line, multi-gpu becomes the only way to improve, and if you want to do something like Surround, you would be wanting multi-gpu to max out 3 high-res monitors.
For a first attempt at a motherboard, it's pretty good.
In time, hopefully they'll produce something special - IIRC the team that's made these motherboards is the team that brought us most of the recent EVGA motherboards.
"I completely agree with (both of) you but my question was why do they recommend to not use a multi card configuration when choosing a GPU but then make it an important factor in choosing a motherboard."
-
« Previous
-
1
-
Next »
Discuss in the forums