- Status
jj463rd
Distinguished
- Apr 9, 2008
- 1,510
- 0
- 19,860
Well as most of us already know the Phenom II X4 955 was never meant to compete with Intel's Core i7 line which is in a league of its own.
Even if one could match the Core i7 920's performance in games by using better graphics cards (which I think is right now close to possible despite this review (although the review was only attempting to make some good points despite its flaws)), in many other applications (video encoding,others etc) the Core i7 would win by a substantial margin so trying to compete with a Core i7 is foolhardy in terms of its raw performance.
Rather as most of us here well know the Phenom II X4 955 was meant to compete with the Core 2 Quad line and in this it's fairly comparable to the Intel Quad Core 9550 except with better memory bandwidth with its IMC.
Therefore as a somewhat budget CPU especially with the combo deals now, one would prefer to use the Phenom II X4 955 for a medium end gaming rig.Maybe consider Toms Core i7 920 medium end SBM rig a higher medium end rig and the Phenom II X4 955 a lower medium end rig.In this area it does well.The Phenom II X4 955 is better off as a sub $1,000 performer gaming rig in which a Core i7 920 system would probably be relegated to using a sub $100 graphics card or less and end up as a non gaming system.Not all PC users are gamers anyways.Also not all gamers are obsessed by the FPS genre.I have played so many sets of dozens of First Person Shooters (FPS) in the past that I am now completely bored with them and have little interest in the new ones.You know FPS are not the only gaming genre out there.Some people prefer flight simulations (combat and/or civilian),spacecraft flight simulations (real or fictional),sub sims,racing games,strategy,RPG's etc.
I saw a chart that showed the value vs performance of various CPU's.The ones that seemed unreasonable were either the very low end CPU's (like the AMD single cores,early defective original Phenoms) or the very high end Intel Core i7's.On the Chart there were many CPU's that performed well for their cost.The Core i7 920,Core 2 Quad 9550,Core 2 Quad 6600,some budget Core 2 Duo's, the Phenom X4 955,940 and X3 720,X2 550 amongst many others were very worthy economically for their performance on the chart.This review just showed that when one spent too much money on their Phenom II X4 955 system that it would be better off to have gone with building a Core i7 920 system instead so I think that this is the gist of the article.The writer didn't have the clairvoyance to know of the AMD combo deal so this changed the cost analysis equation a lot.I think his remark about not complaining when the minimal frame rate reaches 60 frames per second is either absurd or excessive though.I wouldn't complain about a rare drop to 24 frames per second as the minimal frame rate.My complaints about frame rates are when playing a game it turns into a slideshow.
Even if one could match the Core i7 920's performance in games by using better graphics cards (which I think is right now close to possible despite this review (although the review was only attempting to make some good points despite its flaws)), in many other applications (video encoding,others etc) the Core i7 would win by a substantial margin so trying to compete with a Core i7 is foolhardy in terms of its raw performance.
Rather as most of us here well know the Phenom II X4 955 was meant to compete with the Core 2 Quad line and in this it's fairly comparable to the Intel Quad Core 9550 except with better memory bandwidth with its IMC.
Therefore as a somewhat budget CPU especially with the combo deals now, one would prefer to use the Phenom II X4 955 for a medium end gaming rig.Maybe consider Toms Core i7 920 medium end SBM rig a higher medium end rig and the Phenom II X4 955 a lower medium end rig.In this area it does well.The Phenom II X4 955 is better off as a sub $1,000 performer gaming rig in which a Core i7 920 system would probably be relegated to using a sub $100 graphics card or less and end up as a non gaming system.Not all PC users are gamers anyways.Also not all gamers are obsessed by the FPS genre.I have played so many sets of dozens of First Person Shooters (FPS) in the past that I am now completely bored with them and have little interest in the new ones.You know FPS are not the only gaming genre out there.Some people prefer flight simulations (combat and/or civilian),spacecraft flight simulations (real or fictional),sub sims,racing games,strategy,RPG's etc.
I saw a chart that showed the value vs performance of various CPU's.The ones that seemed unreasonable were either the very low end CPU's (like the AMD single cores,early defective original Phenoms) or the very high end Intel Core i7's.On the Chart there were many CPU's that performed well for their cost.The Core i7 920,Core 2 Quad 9550,Core 2 Quad 6600,some budget Core 2 Duo's, the Phenom X4 955,940 and X3 720,X2 550 amongst many others were very worthy economically for their performance on the chart.This review just showed that when one spent too much money on their Phenom II X4 955 system that it would be better off to have gone with building a Core i7 920 system instead so I think that this is the gist of the article.The writer didn't have the clairvoyance to know of the AMD combo deal so this changed the cost analysis equation a lot.I think his remark about not complaining when the minimal frame rate reaches 60 frames per second is either absurd or excessive though.I wouldn't complain about a rare drop to 24 frames per second as the minimal frame rate.My complaints about frame rates are when playing a game it turns into a slideshow.
Cleeve,
Firstly, thanks very much for doing this rewrite (and sorry it took so long for me to get to it and read it).
I would like to echo the sentiments expressed elsewhere in this thread about memory timings. I didn't think that it really would matter much, but I did find quite a bit of information on the web about AMD architecture and the benefit of low latency ram. There does seem to be a sweet spot between the MHz of the ram and the latency settings, going to high in MHz at a sacrificate of tight timings will have an adverse effect as much as too low MHz with loose timings. There's a balance in there.
My point is that I think when comparing AMD vs. Intel, you _should_ take memory latencies and speed into account. You could actually buy lower speed, tighter timed memory and also put that elsewhere in the system. I don't think it's fair to say that in order to do a AMD vs. Intel, you need to keep the same components _on parts that matter to the arcitecture_. I stress that (not to be obnoxious) because there was some debate about using SSD drives, and I really doubt that amd platform would benefit over Intel on this part (but, full disclosure, I've not done a single google search on that question). Memory latencies, on the other hand, I've found a lot of references to this and Intel defintiely favors higher mhz looser bindings in their memory setup than vs. AMD (which appears to be oppostite).
It would be very interesting for you to put this to the test, but I imagine that's a lot of work, and you may not be so inclined. However, if that's the case, and we should rely on what others have reported on other sites, then you should treat the same mem sticks across amd and intel to be favoring one platform or the other. For this reason, that's why I recommend that you SHOULD be using hardware that favors the platform when doing a dollar-for-dollar comparison.
Thanks again for your great work, and I look forward to future articles.
Firstly, thanks very much for doing this rewrite (and sorry it took so long for me to get to it and read it).
I would like to echo the sentiments expressed elsewhere in this thread about memory timings. I didn't think that it really would matter much, but I did find quite a bit of information on the web about AMD architecture and the benefit of low latency ram. There does seem to be a sweet spot between the MHz of the ram and the latency settings, going to high in MHz at a sacrificate of tight timings will have an adverse effect as much as too low MHz with loose timings. There's a balance in there.
My point is that I think when comparing AMD vs. Intel, you _should_ take memory latencies and speed into account. You could actually buy lower speed, tighter timed memory and also put that elsewhere in the system. I don't think it's fair to say that in order to do a AMD vs. Intel, you need to keep the same components _on parts that matter to the arcitecture_. I stress that (not to be obnoxious) because there was some debate about using SSD drives, and I really doubt that amd platform would benefit over Intel on this part (but, full disclosure, I've not done a single google search on that question). Memory latencies, on the other hand, I've found a lot of references to this and Intel defintiely favors higher mhz looser bindings in their memory setup than vs. AMD (which appears to be oppostite).
It would be very interesting for you to put this to the test, but I imagine that's a lot of work, and you may not be so inclined. However, if that's the case, and we should rely on what others have reported on other sites, then you should treat the same mem sticks across amd and intel to be favoring one platform or the other. For this reason, that's why I recommend that you SHOULD be using hardware that favors the platform when doing a dollar-for-dollar comparison.
Thanks again for your great work, and I look forward to future articles.
Erm, sorry, I made some typos up there, let me fix:
[Old]
There does seem to be a sweet spot between the MHz of the ram and the latency settings, going to high in MHz at a sacrificate of tight timings will have an adverse effect as much as too low MHz with loose timings.
[New]
There does seem to be a sweet spot between the Mhz of the ram and the latency settings: going too high in Mhz but sacrificing tight timings will have a negative impact, but too low Mhz with very tight timings will also give sub-par performance.
And...
[Old]
You could actually buy lower speed, tighter timed memory and also put that elsewhere in the system.
[New]
You could actually buy lower speed, tighter timed memory and put the extra money elsewhere into the system (such as gfx).
Sorry for the typos, been away for a litle bit and I'm still on vacation
.
-Cnox
[Old]
There does seem to be a sweet spot between the MHz of the ram and the latency settings, going to high in MHz at a sacrificate of tight timings will have an adverse effect as much as too low MHz with loose timings.
[New]
There does seem to be a sweet spot between the Mhz of the ram and the latency settings: going too high in Mhz but sacrificing tight timings will have a negative impact, but too low Mhz with very tight timings will also give sub-par performance.
And...
[Old]
You could actually buy lower speed, tighter timed memory and also put that elsewhere in the system.
[New]
You could actually buy lower speed, tighter timed memory and put the extra money elsewhere into the system (such as gfx).
Sorry for the typos, been away for a litle bit and I'm still on vacation
.-Cnox
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]ImaxAMD[/nom]Thats about all he has going for him.[/citation]
Well, I do have a nifty job where I get to influence hundreds of thousands of folks a month over an internet publication, which is kind of cool.
So as little as I have going for myself, i can take a bit of solace in the fact that my opinions will inevitably bear a lot more weight than those found, say, on a comments forum... 😀
Well, I do have a nifty job where I get to influence hundreds of thousands of folks a month over an internet publication, which is kind of cool.
So as little as I have going for myself, i can take a bit of solace in the fact that my opinions will inevitably bear a lot more weight than those found, say, on a comments forum... 😀
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]Cleeve[/nom]Average frame rates over 60 aren't overkill when minimum frame rates can dip much lower. If minimum framerates are over 60, then you'd have nothing to complain about.[/citation]
[citation][nom]playerone[/nom]Your moronic thoughts on the left side of your brain have penatrated the right side of your brain where your tech center is. [/citation]
Wow. So to you, 60 fps average is perfectly good even if it dips to 2 fps when things get busy?
Now, I know you can do better than that. You've done pretty well pretending you're not completely incompetent, don't spoil it with obvious misinformation when you're on a roll. 😉
[citation][nom]playerone[/nom]Your moronic thoughts on the left side of your brain have penatrated the right side of your brain where your tech center is. [/citation]
Wow. So to you, 60 fps average is perfectly good even if it dips to 2 fps when things get busy?
Now, I know you can do better than that. You've done pretty well pretending you're not completely incompetent, don't spoil it with obvious misinformation when you're on a roll. 😉
Just built a rig with 955 and waiting for my 4890.
Starting to regret it.. but it would have cost about 175 more for a quad mobo and the 920.
My main decisions for building the AMD system was:
1. Extra dollars will go to another 4890
or wait for the 5000 series (most likely scenario)
2. Just got burned by building my last system intel based. (less than 2 years ago)
Built it to upgrade, but when I'm ready... can't even put the fastest processor in the MOBO... (mobo manufactures fault really.. socket is good but no bios update to support newer processors)
But by jumping on the am3 board early, I hope to get at least one quality upgrade before having to buy an new mobo
Starting to regret it.. but it would have cost about 175 more for a quad mobo and the 920.
My main decisions for building the AMD system was:
1. Extra dollars will go to another 4890
or wait for the 5000 series (most likely scenario)
2. Just got burned by building my last system intel based. (less than 2 years ago)
Built it to upgrade, but when I'm ready... can't even put the fastest processor in the MOBO... (mobo manufactures fault really.. socket is good but no bios update to support newer processors)
But by jumping on the am3 board early, I hope to get at least one quality upgrade before having to buy an new mobo
[citation][nom]Cleeve[/nom]Wow. So to you, 60 fps average is perfectly good even if it dips to 2 fps when things get busy?Now, I know you can do better than that. You've done pretty well pretending you're not completely incompetent, don't spoil it with obvious misinformation when you're on a roll.[/citation]
Juzz shut up Cleevage[citation][nom]Cleeve[/nom]Well, I do have a nifty job where I get to influence hundreds of thousands of folks a month over an internet publication, which is kind of cool.So as little as I have going for myself, i can take a bit of solace in the fact that my opinions will inevitably bear a lot more weight than those found, say, on a comments forum...[/citation]
Funny how at least how this lowly "commenter" hasn't had to take back a single point that you can't frankly refute on any level so Don't forget the part about how you can embarass yourself in front of "Hundreds of thousands" with the most incompetant jiberish that needs an apology the following "Month" after your "influance" has taken people down the wrong path. Dickhead! (Sorry for my bad inglish)
Juzz shut up Cleevage[citation][nom]Cleeve[/nom]Well, I do have a nifty job where I get to influence hundreds of thousands of folks a month over an internet publication, which is kind of cool.So as little as I have going for myself, i can take a bit of solace in the fact that my opinions will inevitably bear a lot more weight than those found, say, on a comments forum...[/citation]
Funny how at least how this lowly "commenter" hasn't had to take back a single point that you can't frankly refute on any level so Don't forget the part about how you can embarass yourself in front of "Hundreds of thousands" with the most incompetant jiberish that needs an apology the following "Month" after your "influance" has taken people down the wrong path. Dickhead! (Sorry for my bad inglish)
[citation][nom]Cleeve[/nom]Wow. So to you, 60 fps average is perfectly good even if it dips to 2 fps when things get busy?Now, I know you can do better than that. You've done pretty well pretending you're not completely incompetent, don't spoil it with obvious misinformation when you're on a roll.[/citation]
Heh, nice try Cleevy. Pretty funny and pretty sure I never said any of that. Here let me try what you have been doing to some of the "commenters" when they show you up with real facts....
Heh, nice try Cleevy. Pretty funny and pretty sure I never said any of that. Here let me try what you have been doing to some of the "commenters" when they show you up with real facts....
[citation][nom]Cleeve[/nom]That's exactly what I said in the conclusion, I suck up to what ever I feel, sometimes that means I act like a fanboy and dont think about what I am doing when I put a system together that I dont feel is worthy. Does this mean we recommend staying away from AMD CPUs? Not at all. But AMD's offerings shine below the Pentium4's price segment. especially when you compare the whitepapers concerning the new DX11 Api's aproach to specular alaiasing overlays, they just performe much better than sprites or alpha textures. Also, even more money can be saved if you're willing to settle for DDR1 with one of AMD's more value-oriented chipsets. If this kind of low-cost system is overclocked, we should see gaming results very similar to the ones we saw in our I7 test today.."[/citation]
Again Cleevy you dont know what you are saying, you think an old DirectX8 command should be double mapped thru a single shader and fed thru older DDR1 and then it beats the I7? What are you freaken saying dude? LoL! Ok there buddy, ur the xpert. Love you Cleevy
Again Cleevy you dont know what you are saying, you think an old DirectX8 command should be double mapped thru a single shader and fed thru older DDR1 and then it beats the I7? What are you freaken saying dude? LoL! Ok there buddy, ur the xpert. Love you Cleevy
[citation][nom]Cleeve[/nom]ImaxAMD, I whish a hundred thousand times a month that someone would read my reviews but Just because I kind of just throw the hardware at the benchmarks is no reason to put my reviews down, they are what they are and there is nothing that you or any of you lowly comment types can do to change that! I'm up here and you are down there.[/citation]
Your a doooushbag Cleevus!
Your a doooushbag Cleevus!
[citation][nom]playerone[/nom]Again Cleevy you dont know what you are saying[/citation]
Well, since you made up that bogus quote, I'll agree I don't know what you've *made up* that I've said.
You're even easier to manipulate than ImaxAMD. Instead of throwing personal insults, you've lowered yourself to making up quotes. Heheh... No actual arguents left, huh?
All too easy. 😀
Well, since you made up that bogus quote, I'll agree I don't know what you've *made up* that I've said.
You're even easier to manipulate than ImaxAMD. Instead of throwing personal insults, you've lowered yourself to making up quotes. Heheh... No actual arguents left, huh?
All too easy. 😀
pauldh
Illustrious
- Jan 1, 2004
- 11,814
- 0
- 40,810
Sure, I understand what you are saying. Just keep in mind, this was a specific comparison between two popular high-performance CPU's, not an overall AMD/Intel value shootout. People(including the author himself) bring up good points about using less expensive AMD CPU's for games, but exploring those is outside the scope of this story.
To be clear, the price point was not chosen to to favor the i7. Multi/GPU's make sense at this level of CPU, and help eliminate GPU limitations. The same author of the Cyperpower story previously did the $1300 SBM Micro ATX machine. He just wanted to take the opportunity to explore AMD"s best CPU vs. that i7 920 build. To be honest though, had he used a DDR2 AM2+ mobo to knock down cost (adding value) there would then be those would would cry foul that Intel paid him not to use AM3/DDR3 (for the slight performance bump).
Just keep this in mind, that both Don and myself (author of $600 SBM PC) have stated both in articles and comments that AMD looks very good below that $1250 or so price range. i7 + $300 or more in Graphics is a potent gamer and all round machine. i7 really shines in some of our applications too, hence he used in for the SBM. But a sweet way to bring that system cost down without any large sacrifice in gaming performance would be a X3 720/ X4 940 DDR2 and the same(similar) graphics. I haven't seen anyone recommend i7 + single $100-150 card for gaming. But forgetting best bang/buck... if i7 + SLI/Crossfire $150+ cards is an option within budget, it's hard to argue that is a bad build, Right?
And yes you are right, when I said To Others: and mentioned the OC'in, I was not talking to you. I didn't anticipate sticking around the discussion much beyond that one post. (somehow that changed) :lol:
Have a nice weekend sreams.
The northbridge overclock has been found by another hardware site to give about 10% +fps in a similar configuation. Additionally, testing at 2560 x 1600 would likely give more of an advantage to the higher clocked GPUs.
Don't get me wrong; it does seem the i7 is a must at this price point (given that anyone paying 1k+ for a monitor will not save 200 on the CPU/mobo/memory to get bottlenecked with a 600$ system otherwise), but it may be that it's unnecessary (i.e. waste of money) at a resolution of 1920 x 1200 and 60 FPS on most current monitors. Without testing at 2560 x 1600, do you really see anything to make 2 x 4890 more value for money than 2 x 4870, regardless of processor / mobo? If you simply cut all benchmarks to a max of 60 fps, would anything except the Crysis benchmarks (that did show a win for the Phenom) really matter at these resolutions?
The most overclockable / game performing AMD CPU is currently (claimed to be by the most recent review I read) the 720 X3. If you give that one a decent mobo/memory, would it not give you the same performance w dual 4870s as an i7 system at a far lower price point?
(Not an AMD fanboi here; I just saw the feedback on the latest SBMs and read some stuff on percieved Tom's Intel bias via Google. You do need to try out a decent Phenom II build to get rid of the bias allegations. Presumably the 600 and 1300 price points are just too low/high to make an AMD build the best, but most real computers fall between these price points.)
Don't get me wrong; it does seem the i7 is a must at this price point (given that anyone paying 1k+ for a monitor will not save 200 on the CPU/mobo/memory to get bottlenecked with a 600$ system otherwise), but it may be that it's unnecessary (i.e. waste of money) at a resolution of 1920 x 1200 and 60 FPS on most current monitors. Without testing at 2560 x 1600, do you really see anything to make 2 x 4890 more value for money than 2 x 4870, regardless of processor / mobo? If you simply cut all benchmarks to a max of 60 fps, would anything except the Crysis benchmarks (that did show a win for the Phenom) really matter at these resolutions?
The most overclockable / game performing AMD CPU is currently (claimed to be by the most recent review I read) the 720 X3. If you give that one a decent mobo/memory, would it not give you the same performance w dual 4870s as an i7 system at a far lower price point?
(Not an AMD fanboi here; I just saw the feedback on the latest SBMs and read some stuff on percieved Tom's Intel bias via Google. You do need to try out a decent Phenom II build to get rid of the bias allegations. Presumably the 600 and 1300 price points are just too low/high to make an AMD build the best, but most real computers fall between these price points.)
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]dsjhkgfdsjhgdh[/nom]You do need to try out a decent Phenom II build to get rid of the bias allegations. [/citation]
Not quite; at this point, the Phenom II had the advantage with more push on the overclock than the i7.
As it stands, these are two systems put together with a basic overclock. To be fair, if I went back to the drawing board to optimize the Phenom II overclock, then I'd also have to go back to the drawing board to optimize the i7 overclock. And we all know an i7 can go a hell of a lot higher than 3.45 GHz.
Yet, if I optimize the Phenom II and then optimize the i7, we know what will happen. The same few voices will cry foul.
I'm quite satisfied with the testing, results, and conclusion. I'm not going to intentionally write up a biased revision just to make a couple complainers feel better about themselves.
Not quite; at this point, the Phenom II had the advantage with more push on the overclock than the i7.
As it stands, these are two systems put together with a basic overclock. To be fair, if I went back to the drawing board to optimize the Phenom II overclock, then I'd also have to go back to the drawing board to optimize the i7 overclock. And we all know an i7 can go a hell of a lot higher than 3.45 GHz.
Yet, if I optimize the Phenom II and then optimize the i7, we know what will happen. The same few voices will cry foul.
I'm quite satisfied with the testing, results, and conclusion. I'm not going to intentionally write up a biased revision just to make a couple complainers feel better about themselves.
I don't doubt the i7 is strictly better at this this price point; I meant Tom's Hardware needs to do a decent AMD build to get rid of the cries of foul, rather than meaning you specifically need to change this review. I think people complain so much that you (again meaning the site) should take notice and try to address their concerns.
The conclusions are fine IMO, but I don't think the price point is anywhere near optimal for an AMD build given the emphasis on OC and gaming performance. It answers the question of whether an 1300 AMD can beat an 1300 i7 with the added GPU power which is simply answered with "No, most realistic tasks are CPU limited when you have this much GPU power".
Fair OCing in reviews are obv difficult. You just need to pick a point where it seems many systems could run with little knowledge/effort. I thought the NB OC was worth mentioning for future reference, so you could try it out and possibly educate the readers.
Anyway, I've already picked the q9550s for the only build I'll make in the next few years. I just thought it worth mentioning that it seems AMD might be competitive in the segment between your 600 and 1300 builds _and_ 60 FPS @ 1920 x 1200 is the max most are aiming for unless they have 30" monitors.
The conclusions are fine IMO, but I don't think the price point is anywhere near optimal for an AMD build given the emphasis on OC and gaming performance. It answers the question of whether an 1300 AMD can beat an 1300 i7 with the added GPU power which is simply answered with "No, most realistic tasks are CPU limited when you have this much GPU power".
Fair OCing in reviews are obv difficult. You just need to pick a point where it seems many systems could run with little knowledge/effort. I thought the NB OC was worth mentioning for future reference, so you could try it out and possibly educate the readers.
Anyway, I've already picked the q9550s for the only build I'll make in the next few years. I just thought it worth mentioning that it seems AMD might be competitive in the segment between your 600 and 1300 builds _and_ 60 FPS @ 1920 x 1200 is the max most are aiming for unless they have 30" monitors.
thebatman
Distinguished
- Jun 30, 2009
- 15
- 0
- 18,510
[citation][nom]Cleeve[/nom]Not quite; at this point, the Phenom II had the advantage with more push on the overclock than the i7.As it stands, these are two systems put together with a basic overclock. To be fair, if I went back to the drawing board to optimize the Phenom II overclock, then I'd also have to go back to the drawing board to optimize the i7 overclock. And we all know an i7 can go a hell of a lot higher than 3.45 GHz.Yet, if I optimize the Phenom II and then optimize the i7, we know what will happen. The same few voices will cry foul.I'm quite satisfied with the testing, results, and conclusion. I'm not going to intentionally write up a biased revision just to make a couple complainers feel better about themselves.[/citation]
Lol man, if I had a dollar for everytime you've had to point out this clearly obvious fact....
Lol man, if I had a dollar for everytime you've had to point out this clearly obvious fact....
- Mar 7, 2003
- 12,012
- 0
- 40,790
Well if that's what you mean, I'll agree. I admitted straight up this wasn't an ideal setting for a review, pressed for time and inserted into the schedule to address most of the concerns of a previous review. Having said that, I'm more than happy with the results and conclusion all things considered.
But are you asking if we'll try harder next time? Sure, why wouldn't we? Of course, it's a given that to keep things on an even keel we have to try harder with all the competitors. I'm not going to spend 12 hours tweaking the next PhII build and then take the first i7 overclock that I can get... that'd be showing flagrant favoritism.
But I'm not deaf, I hear and acknowledge the concerns. I might disagree as to their relevancy in this case, but I can accommodate. I'll blast trolls who feel the need to belittle others to make themselves feel better, but I'm not going to ignore genuine concerns from honorable forum members. That wouldn't make any sense.
astrodudepsu
Distinguished
- Jun 11, 2009
- 340
- 0
- 18,790
Kudos mate. Glad to see this topic done right. The AMD people are starting to rival apple folks in terms of sheer zealotry. It is EXTREMELY obvious to anyone with a functioning brain that if anything this test was favored as much towards AMD as is possible while still maintaining 'fairness'. I am delighted to see that AMD is once again a viable alternative with their P2 line. I've bounced back and forth between intel and AMD since I got into this hobby (only) 8 years or so ago. Viable competition is great for consumers, I wish more people would understand this.
not a fair test, a fair test is everything being equal, right?
1st the video cards are different.
2nd the harddrives are different, some might say "so" but if u factor in that its 1.5 tb harddrive to a 640 gb. now if u have ever defragged a 60gb hd 7200rpm, to a 160gb hd 7200rpm w/equal cache, w/equal stored, the 60gb will be 1st to be finished. why u ask? because of the vast storage area difference that has to be accessed. yes it does search the whole disk space available everytime, this might take a few milliseconds to a few seconds. and that is a big difference in a game bench mark.
also, the 1.5 tb hd has a 16mb cache, and the 640gb hd has double at 32mb cache, this is a huge difference.
lastly, all gaming bench marks are played at 0 AA or 4x AA, well maybe 2 at 8x AA, the AF setting mostly at 4x or none. u say so what? well if im going to lay down 500+ for 2 video cards i want that game running with AA and AF set to maximum. and this is where the cards would shine, ATI cards for years now are known for this.
p.s. can anyone find a 4870x2 - 4890x2 comparison chart? ive seen in some occasions that the 4870x1 beat the 4890x1. ill find the articles.
1st the video cards are different.
2nd the harddrives are different, some might say "so" but if u factor in that its 1.5 tb harddrive to a 640 gb. now if u have ever defragged a 60gb hd 7200rpm, to a 160gb hd 7200rpm w/equal cache, w/equal stored, the 60gb will be 1st to be finished. why u ask? because of the vast storage area difference that has to be accessed. yes it does search the whole disk space available everytime, this might take a few milliseconds to a few seconds. and that is a big difference in a game bench mark.
also, the 1.5 tb hd has a 16mb cache, and the 640gb hd has double at 32mb cache, this is a huge difference.
lastly, all gaming bench marks are played at 0 AA or 4x AA, well maybe 2 at 8x AA, the AF setting mostly at 4x or none. u say so what? well if im going to lay down 500+ for 2 video cards i want that game running with AA and AF set to maximum. and this is where the cards would shine, ATI cards for years now are known for this.
p.s. can anyone find a 4870x2 - 4890x2 comparison chart? ive seen in some occasions that the 4870x1 beat the 4890x1. ill find the articles.
sry for double post,
also, the 1.5 tb hd has a 16mb cache, and the 640gb hd has double at 32mb cache, this is a huge difference. ever run vista rig w/1gb ram, then on a rig w/2gb? even if the rigs have different setups the 1gb ram 1 will be sluggish if at all usable, the 2gb will be smooth.
also, the 1.5 tb hd has a 16mb cache, and the 640gb hd has double at 32mb cache, this is a huge difference. ever run vista rig w/1gb ram, then on a rig w/2gb? even if the rigs have different setups the 1gb ram 1 will be sluggish if at all usable, the 2gb will be smooth.
astrodudepsu
Distinguished
- Jun 11, 2009
- 340
- 0
- 18,790
this was about an equal PRICE RANGE, not equal parts. If you placed two 4890s with an i7 the AMD solution would have lost by a greater margin.
The HDD complaint is trivial at best.
Both rigs used ATI cards, so I'm not sure where your comments on AA and AF have any merit whatsoever. There were no nvidia cards, so if they would have jacked up AA and AF they would have scaled very similarly.
The HDD complaint is trivial at best.
Both rigs used ATI cards, so I'm not sure where your comments on AA and AF have any merit whatsoever. There were no nvidia cards, so if they would have jacked up AA and AF they would have scaled very similarly.
- Status
TRENDING THREADS
-
News UserBenchmark bashes AMD GPUs and claims they lack real-world performance- Started by Admin
- Replies: 128
-
Review AMD Ryzen 9 9950X3D Review: Stunning gaming performance meets top-tier productivity
- Started by Admin
- Replies: 69
-
-
-
-
Review AMD Radeon RX 9070 XT and RX 9070 review: An excellent value, if supply is good
- Started by Admin
- Replies: 169
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
Facebook
Twitter
Instagram