Cyberpower’s Gamer Dragon: Can AMD Bring The Game?

Page 9 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Please explain (if you don't mind) the closing as it was flame bait. As I stated the testing was fine the PC is a POS. You did not compare this cyberBS system in the closing you clearly and intentionality made an AMD v/s Intel Fanboi statement. You can not misrepresent the facts and that is where most people are having a problem. THIS IS A CRAP SYSTEM BUT YOU CLEARLY MISLEAD THE GENERAL PUBLIC BY LETTING THIS REVIEW REPRESENT AMD AND NOT JUST CYBERPOWER. Read your own post and then read the conclusion and stop lying to yourself. You yourself said it was a POS therfore your closing was Fanboi BS.

[citation][nom]Cleeve[/nom]Oh for chrissake. It's always the same old accusations of being paid off when things don't go like fanboys expect them to, isn't it?Look, there's no bias happening here. I tested these machines without any expectation either way, frankly I was just as surprised at the results as you guys are. But I'm not going to hide what I've found to satify people who have a hard-on for one brand over the other.Half the time I'm getting accused of bias from AMD fanboys, the other half it's Nvidia fanboys, and now it's the same thing with intel and AMD. Look, I don't give a flying rat's bunghole who is better, I test it, report on it, and form opinions based on the testing.Now, with these results I'm not sure how you could spin it any other way. Two 4890's should be more powerful than two GTX 260's, so if the stock i7 is dominating the overclocked Phenom II then I don't really have a choice but to come to the obvious conlusion.Was there an anomaly with the Cyberpower system? Look, I performed a lot of these tests twice because the results were so dramatic. I was as surprised as any of you. It seems to be running fine, no crashes. I ran the tests with crossfire disabled and got even lower framerates, so I'm pretty sure crossfire was working.Some of you see this article for what it is, a test of a sample Phenom II and a comparison to a core i7 system. To those of you I appreciate the support. To the rest of you... you want to read an anti-AMD conspiracy into it? Have fun. I've been recommending AMD cards for months in the Best Cards for the Money article.But I tell you what... just for you suspicious types, I'm going to put together tests with a Core i7 vs. a Phenom II, using the same 4890's in crossfire. Same RAM, to😵f course, if the Intel still has a leg up, I'll hear nothing but cries of foul play. But that comes with the territory when dealing with fanboys in any camp. At the very least, I'll know I did absolutely everything I could to put an end to the ludicrous accusations of being paid off.[/citation]
 


Yeah Archer, I've kind of responded to comments in the same vein as yours a few times now.

Check the previous 11 pages or so for a response you find suitable, and just apply it to your question. Then you're good to go, and I don't have to waste any time.
 
Can you or I help that it is a crap system all I would like to have seen is a closing that did not compare this system to a clearly superior system in most respects. I really don't think you had intent but that is how it sounded. I really dont have to read the other pages again and I know the hassles and someone will always be pissed no matter what. I rather thought the reviev was good without that one paragraph.
 
Well what cleeve is going to put up is going to prove him right, maybe with a bit lesser margines than in this artichel. Put 2 4890 in an overcloked i7 920 vs an overclocked phii 955 and intel take all the victories, only way Amd can get a a draw is bye using settings that stress 2 4890 so they become the bottelneck. This is proven bye all the major computersites in diffrent countries, so i cant see that there are anything left there to prove.

But a right set up AMD rig should stand a chance against the SMB build. This is due to the fact that u have round 380 us dollars to spend on other aquipment than cpu, motherboard and memory. Among the best an AMD build can put in is an vertex SSD disk, and better grapich cards.

But with same equipement and when price is not an issue, the result is given.
 
Whatwould be a real challenge for anyone and cleeve, is can we build an AMD rig that can give the SMB a run for its money, that would be fun to see. No sites have ever done builts like that, and that sure would prove that Toms isnt bought bye anyone.

Well that aside, i been reading all computerites since the beginning of the 90'. And well Toms have been know to tweek their test so Intel would win , even when it was prescot vs athlon, Toms was one of the few sites that found that Intel was best for evrything or had no margins at all. But i have actually read sevral articles on here that slowly has given the creditbility back to this site.

But denying that i7 is the best cpu out there, that is pointless, however up to a pricepoint u can always build a better overall rig due to the priceadvantage of the phII. Becous a rig performance isnt soly the cpu's performance.
 
[citation][nom]osse[/nom]Whatwould be a real challenge for anyone and cleeve, is can we build an AMD rig that can give the SMB a run for its money, that would be fun to see. No sites have ever done builts like that, and that sure would prove that Toms isnt bought bye anyone.Well that aside, i been reading all computerites since the beginning of the 90'. And well Toms have been know to tweek their test so Intel would win , even when it was prescot vs athlon, Toms was one of the few sites that found that Intel was best for evrything or had no margins at all. But i have actually read sevral articles on here that slowly has given the creditbility back to this site.But denying that i7 is the best cpu out there, that is pointless, however up to a pricepoint u can always build a better overall rig due to the priceadvantage of the phII. Becous a rig performance isnt soly the cpu's performance.[/citation]

I have no doubt that the outcome would be the same but the numbers would give a decent AMD setup a better showing.

This is the first time I have had any real problems with articles from Toms. In writing a review the point is to present the facts and draw a conclusion from those facts and this was done. It is not however to steal the spotlight from the product being reviewed (let it die on it's own merits).

I think the conclusion was a little misplaced and some of reviewers comments in the conclusion seem to be brought from frustration and trying to justify reasons for the i7 that was used earlier for SBM and flack was thrown at him why not return fire to vindicate himself? It makes the site look bad personally I say trash the attacks as they come don't carry it over into the rest of your reviews.
 
Well in pure number crunshing I7 would win. Intel x25 is the fastest disk round, but even an vertex SSD do improve lots of aspect in contrast to a regular 7200 rpm disk. And u can affored an vertex and still be within the same price as the SMB build.

http://www.anandtech.com/storage/showdoc.aspx?i=3403&p=13

What did supprise me most was how good 2 x 260gtx performed. But as i said a goodgamerigg shell be build for turning settings on max or most possibel and that with resulution 1920x1200, or are there some that actually buy a rigg in that pricesegment and run it on dirt sheap lcd screen that cant handel that reolutin, for heavnes sake, in 1280x800, a rigg like this is a total waste. Maybe some run a rigg like thouse in the test on 1650x1050, but even that is a waste.

As i metion only real high end screens can display more than 60 fps. So to test 2 rigs vs eachother, u have to crank the settings up as far as t hey go for evry game, and still get close to 60 fps. Only benchers care if u get 100 + frames with low settings.

Hardcop does that (when they test grapic cards), and its the only way to tell wich rigg is the best gamerigg. Find out what settings that allows good gameplay , pure and simple
 
And when it comes to the 260 gtx vs 4890, it also shown in test after test, while the 260 suffer a lot when cranking up the AA and other settings the 4870 and 4890, does not.

So in a way this review was a cpu review "shrugs", and even an AMD fantaic like wouldnt say that I7 isnt the best cpu, and it shines when u get CF with 4870 - 260 gtx and better.
 
I could build a system that would beat the snot out of both rigs. And guess what? Its AMD & ATI baby.
 
Cleeve,
It's very much appreciated that you want to re-run the benchmarks, but I just want to try to catch you before a whole new brand of firestorm erupts: the initial point of this article (I thought) was that this was a Dragon platform compared to an i7 setup. I thought that is where AMD was trying to compete: they have a compelling product if you use the platform (mainboard/CPU/GPU), get the drivers, have the system controlled overclocking and have it shutdown services etc..you know all the thigns they were touting about the dragon platform...

This includes the use of specific video cards (4800's or 4900's i forget) but going to the 260s in my mind is _not_ the dragon platform. And the other point of going with an AMD config is that you sacrifice CPU to bulk up on GPU. By going with the same GPUs across the board you are AGAIN (intentionally or not) pushing the competition back to a straight up CPU vs. CPU fight, and that is NOT the point, the point is putting your money into the system where it counts. Most people I think believe that GPU is where the most bang for buck can be realized. You could put that to the test in this review, but not if you start making it all about the CPU.

So, just trying to stop what I think will be a moot test...everyone agrees that i7s are more optimized and efficient, and you pay for that. The question is, with the Dragon Platform, do the platform components come togethe rin such a way that for the same price, you get better gaming performance? And, I relize the quesiton is slightly loaded, but for me, gaming is the only thing that matters on my pc.

-C
 
[citation][nom]Cnox[/nom] By going with the same GPUs across the board you are AGAIN (intentionally or not) pushing the competition back to a straight up CPU vs. CPU fight, and that is NOT the point, the point is putting your money into the system where it counts. [/citation]

Dude, I totally agree with you. But people have called foul, blaming Nvidia optimizations in these games - which I think doesn't stand up as IMHO the GTX 260 is notably less powerful than the 4890.

In any case, now I'm doing a straight-up comparison. But I like the way you think, so for shits and giggles I might under-clock the 4890's to 4870 speeds in the i7 system to simulate the price disparity for a test or two.
 
Good to see a new comparison will be happening.

The majority here, I hope, agree that the core i7 beats the phenom II. It is simply a better processor.

What we hopefully understand that's been called in to question is the reason given for the rather large differences in the test results. We know that the i7 should beat the Phenom II, especially in cases where the CPU is the bottleneck. What I, and I think most others, disagree with is that the disparity is this significant. The article here: http://www.hardocp.com/article.html?art=MTY0NCwxLCxoZW50aHVzaWFzdA - which I linked earlier, shows that with a GPU that is significantly less restraining, the Core i7 simply outpaces the phenom 720 and 810 CPUs at similar clock speeds. But again, the differences are not nearly as significant as with the Cyberpower vs SBM systems. It'd be interesting to see how a 955 would hold up, maybe we'll get an idea with the new article.
 
Any particular reason we can't just have 6Gigs in each system? I have noticed a few areas in Crysis for example where the RAM will jump over 5 Gigs...but I bothered to stick 8Gigs in both my gaming rigs for this reason. I've even seen UT3 do it if I play long enough.
 
810 X4 @ 3.6ghz
4870 X2
2560x1900 res, 2xAA ,16AF
Highest ingame settings.

Average 74fps.

http://www.hardocp.com/article.html?art=MTY0NCw2LCxoZW50aHVzaWFzdA==

What is the explanation on why this 3.6ghz 955 BE cpu, with two superior graphics cards only gets 56-62fps average at a lower resolution?
 
Here's what I've come up with after much thoghut about the point of this review, and thanks to Cnox for reminding me.

The question originally raised in the SBM, and remains from the Cyberpower article: Does the money saved on the Phenom II allow for a better all-round gaming rig compared to a similarly priced i7?

I did a little newegg shopping. While stuff like cases, power supplies, hard drives, etc. can be swapped between a Phenom II or i7 for the most part, the core components that don't swap out are the CPU, memory, RAM, and CPU cooler.

Spec'ing out two low priced systems concentrating on these components, here's some numbers:

Intel:
Core i7 920 279.99
G.Skill 6GB PC3 10666 94.99
Mobo: DFI X58T3H6 199.99 (Cheapest: MSI X58M @ $169.99 - $30 delta)
Cooler: Xigmatek Drkngt 39.98
----
TOTAL i7 CORE COMPONENTS: $614.95

AMD:
Phenom II 955 245.00
G.Skill 4GB PC3 10666 64.99
Mobo: M4A79T Deluxe 179.99 (cheapest: DFI Lanparty DK 790FX-M2RS @ $134.99 - $45 delta)
Cooler: Zerotherm NV120 49.99
----
TOTAL PhenomII CORE COMPONENTS: $539.97

This leaves us with a $75 delta between system cost when using a Phenom II 955 or i7 920 based system. Even if we use the cheapest motherboards available for either platform, we're looking at a $90 difference in system cost.

Coincidentally, the difference in price between a pair of Radeon HD 4890's and a pair of Radeon HD 4870's is $80. Just about perfect.

So, we'll be testing the i7 machine with a pair of 4870's, and the Phenom II with a pair of 4890s. This douses any potential Nvidia/Ati video card issues, and gives the Phenom a real-world scenario where it has the best chance to stand against the i7.

 
Ah, the hardocp test used quadfire 4870's.

However, surely they should be seeing the same bottleneck as this system? More powerful cards won't matter if it is like the author of this article is suggesting - that the 955 BE is bottlenecking the 4890 crossfire.

Much more likely is this - there is another bottleneck in the Cyberpower setup, which has to be the motherboard in all likelyhood.

Hardocp's farcry 2 benchmarks prove without any doubt that even an 810 X4 Phenom II doesn't bottleneck a far superior graphics setup. Something else is badly wrong with the Cyberpower system, and it most certainly is not the 955 BE.
 
[citation][nom]jennyh[/nom]What is the explanation on why this 3.6ghz 955 BE cpu, with two superior graphics cards only gets 56-62fps average at a lower resolution?[/citation]

The explanation would be HardOCP isn't using the built-in Far Cry 2 benchmark like we did. They used their own custom savegame.

Different benchmark = different results.
 
[citation][nom]da bahstid[/nom]Any particular reason we can't just have 6Gigs in each system? [/citation]

Two reasons:

1. If you use 6GB in the Phenom II system, you have to use three memory slots. Since you have to use an even number of slots for sual-channel memory, you force the motherboard to run in single-channel mode.

2. Using the same amount of memory almost erases the Phenom II's cost advantage as I've layed it out. Part of what makes the Phenom II attractive is it's cost benifit ratio.
 
[citation][nom]Cleeve[/nom]The explanation would be HardOCP isn't using the built-in Far Cry 2 benchmark like we did. They used their own custom savegame.Different benchmark = different results.[/citation]

How can you explain the 810 Phenom II *not* being a bottleneck for a much superior graphics setup in the Quadfire 4870's then?
 
The setup u just put up for the AMD rig cleeve, well no builder with knowlage would say that is the most cost effective solution, the most costeffective is phii940 ddr2 ram, 790fx am2+ motherbored, and vertex SSD drive, that do atleast here match the excat cost as the SBN build.

DDR3 gives round 1-3% effekt over DDR for an phenom system, small cost diffrense

790FX gives round 3-5 % better in games than the cheaper chipsets.

There is a pricediffrense at round 65 us $ betwee 790 FX at am2+ and am3.

And there is littel diffrense between a phii940 and phii 955 clocked at 3.8. There are maybe a bit more headroom in a phii 955, but using it in a build off is like changing from i7 920 - i7 940, u gain ektreamly littel for paying a lot more.

So in my opinion no serious AMD builder would put that up against ur SBM build. A good builder would find room for an vertex, and there is room for that , at least as prices are here.
 
And when it comes to memory, using 8 gb is not a bad idea for a builder, thats allow to turn off disk swap file, wich also reduses framedrops
 
[citation][nom]Cleeve[/nom]Well, I'm more concerned about the 4GB vs. 6GB memory advantage of the i7 than the dual-channel vs. triple channel RAM problem. Not sure how to get around this one.Do you guys think 4GB vs. 6GB is more fair?If I could put two 3GB sticks in the Phenom II system, that would be ideal... anybody know where I can pick up 3GB sticks of DDR3?[/citation]

I say try this, wouldn't cripple the i7 in any what and from what I remember, there is no bonus over 6gb. on 90% of any benchmark

http://www.tomshardware.com/reviews/memory-module-upgrade,2264.html

i7=6gb

test the 955 with synthetic (and possibly one game) with 4gb and 8gb to show if there is any difference to be had (speculating it will be less than 1% as it was using an i7 with 3gb vs 6gb vs 12gb)

Show that there is no virtual difference and prodeed with the testing, wheter you go with 8gb or 4gb on the 955 system.



 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom