Intel Conroe VS. AMD: Biased Tests

Status
Not open for further replies.

Excelsius

Baseband Member
Messages
81
I was interested in the new Conroe and decided to find out more about it. I went through many different forums and eventually realized that a lot of people believe the test of Conroe vs. AMD was biased towards Intel because both the system and the benchmarks were set up by Intel that would optimize it (e.g. the Intel motherboard). The original test was done by Hexus.com.
I thought that maybe there is some truth to all this and perhaps no one would be able to tell definitively how well Conroe performs until the test is performed by an unbiased party. If you really think about it, all the examiners did was to push some buttons on a computer that was set up by Intel. Anyway, I just came across this website: http://sharikou.blogspot.com/2006/04/conroe-performance-claim-being-busted.html I really think that before anyone makes any statements about Conroe he/she should really take a look at these results, which shows that Conroe has large swings in performance in various area.
Additionally I read that Conroe might be the first chip that has been modified to work well with ONE operating system. This means that while Conroe might be great for Windows XP/Vista, it might not work as well for Mac Tiger and their upcoming Leopard (which seems to have exponentially increased software compatibility). What this means is that you might be forced to upgrade your PC every time a new OS comes out (whether Windows or Mac).
 
Very intresting, but they didn't post specs on the systems used for the benchmarking. Can't fully believe them if they don't tell us what setups they used. It's kind of a no-brainer that intel used an optimized setup for conroe, but we still can't believe the hype either way.
 
To start off, I want to congratulate you on finding that! I've seen it, ofcourse, but for someone to find that while doing research...some nice research skills there :)

HOWEVER, that guy is full of crap. Just reading his comments totally disproves his 'facts.' So now that we have that out of the way, the benchmarks that were posted on Anandtech and Hexus were indeed set up by Intel. When the websites got wind of the rumors, Anandtech ran the benchmarks again; this time, by themselves. They also explained all the reasons that they used an older chipset and all that stuff.

The results from Anandtech's revised benchmarks? Same, or better, performance.

But no one really uses the Anandtech benchmarks as the end-all comparison. Most people who actually know a grain's worth about Conroe visit XtremeSystems Forums. The members at XtremeSystems have obtained Conroe pre-release samples. Mainly, they overclock them (A Conroe was overclocked to 4.7Ghz yesterday), but some of them have compared them to other record-setting Athlon 64 processors. Thats where we get most of our information, from non-Intel, fully user-controlled benchmarks from the best overclockers on the planet.
 
I'm just waiting for Conroe and AM2 to be released and then I'll wait and look at the benchmarks. You can't buy it until they are released, and then shortly after that compared by any person and/or sites, so I'll just wait until then.
 
yeah, i would definately agree with GR, they have employes(spelling?) from both AMD and Intel who get pre-release things to test. yes they are the best oc'ers on the planet, maybe in the universe, but i remember a guy with liquid nitro cooling got a P4 w/cedar mill up to 7.0ghz, hopefully they'll be able to get a conroe running like that and have it stable enough to run benchmarks on it.

could you give me the link to the 4.7ghz thread on XS?
 
They aren't Intel employees (if they were, no one would believe them). They're regular people. Well...as regular as you can be while being the best sponsored overclocker on the planet.

Yea, they're good. But you only need to be that good when you're setting world records; when every extra Mhz of speed can make or break a record. If you're planning on taking a 2.13Ghz to 2.9Ghz, thats something that anyone that knows overclocking can do.

4.7Ghz Conroe. 10.750s SuperPi 1M :) I give it 2 more days before we see sub-10s SuperPi 1Ms.
 
Unfortunately the XS website has many problems: many of the links either don't work or time out before loading (I have DSL). I tried to look at the AM2 test results at XS, but the link once again didn't work. I know that Conroe has been compared to the old AMD, but although I haven't read anything about it, I assume that the results were similar when compared to the new AMD AM2.
All this OC discussion, expecially the P IV 7.0 GHz, makes me think that our limit for CPUs is almost exclusively the temperature. Since Conroe is going to be available to retailers I believe on June 4th, I think that it would be very important that the people who conduct the tests obtain it right then and give us all the detailed information before July. It will be interesting to know what would be the comfortable OC limit of the Conroe 2.66GHz so that everyone in this forum could easily achieve (assuming good moral character).
 
Status
Not open for further replies.
Back
Top Bottom