Dual G5 Vs Dual Opteron Benchmark...


Recommended Posts

Has this guy never seen a professional benchmark in his life? We know what CPUs he used, what video card, and what Hard disks. How about some useful information (IE: why, if he was trying to get similar specs did the opteron run ATA-100 rather than SATA?)

What settings were his UT2003 benchmarks run at? How about cinebench and Q3?

What brand of memory?

Which motherboard was in the opteron?

How were the tests run "Best of three", "average of five", "Try it once and use whatever pushes my point"?

There are plenty of good benchmarks posted around the internet and this guy still can't figure out what information we need to evaluate his data. It's almost insulting that he thinks I'm both going to take his word and put value into his claims that a "mystery config" opteron, running benchmarks with apparently random data/settings was beaten by Apple's best offering. Even if he's not lying, there isn't enough information to make this valueable to me.

Barefeats was so embarrasingly pro-mac that if he were in the same room as Microsoft's #1 fan boy (whose initials are PT) that a vortex would open up and swallow the world!"

Link to comment
Share on other sites

Has this guy never seen a professional benchmark in his life? We know what CPUs he used, what video card, and what Hard disks. How about some useful information (IE: why, if he was trying to get similar specs did the opteron run ATA-100 rather than SATA?)

What settings were his UT2003 benchmarks run at? How about cinebench and Q3?

What brand of memory?

Which motherboard was in the opteron?

How were the tests run "Best of three", "average of five", "Try it once and use whatever pushes my point"?

There are plenty of good benchmarks posted around the internet and this guy still can't figure out what information we need to evaluate his data. It's almost insulting that he thinks I'm both going to take his word and put value into his claims that a "mystery config" opteron, running benchmarks with apparently random data/settings was beaten by Apple's best offering. Even if he's not lying, there isn't enough information to make this valueable to me.

Barefeats was so embarrasingly pro-mac that if he were in the same room as Microsoft's #1 fan boy (whose initials are PT) that a vortex would open up and swallow the world!"

it's not really all that much of a mystery, most of the specs are on the second page, but even the people who let him use the comp don't say much about ram speed, mobo, etc.

Link to comment
Share on other sites

We need specifics. Manufacturer and part number. We all know there can be huge differences in performance between manufacturers - even video cards using the same GPU don't preform equally well.

Link to comment
Share on other sites

We need specifics. Manufacturer and part number. We all know there can be huge differences in performance between manufacturers - even video cards using the same GPU don't preform equally well.

I don't know if we have to go that far. Just go for reasonable effort. For the Apple, get the machine and gfx card from Apple and the Ram from Crucial, and for the PC, get the most common model with the same GPU of the same quality class and ram from crucial.

We don't really care what the maximum theoretical speed is but rather what the relative speeds a consumer would get if they bought the machines from a dealer or online store. Most consumers are not going to bother searching out the best components.

Link to comment
Share on other sites

I agree we don't need maximum speed. We do need to know what hardware was compared (was this a best or worst case scenario?) and we need to know what settings were used in all software. We need enough information so that should we be so inclined we can purchase the same hardware, install the same software, and run identical tests. The purpose being so we can either confirm his results (by coming up with similar numbers) or prove him a fraud (by getting vastly different results one way or another).

Don't get me wrong, the G5 is an excellent machine and it may well be faster than some/all dual opteron systems - before I make that claim I'd like to see some unbiased (read barefeats: he's not exactly ambiguous) and/or repeatable tests. Despite all the crap Apple took for their benchmarks when the G5s were released they were at least well documented, done by an unbiased 3rd party, and repeatable. I don't think it's unfair to hold all benchmarks to the same standards if we plan to put any value on them.

Link to comment
Share on other sites

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.