Originally posted by: XBoxLPU
Umm Apple test shows 1024X768, tom's is 640X480
SHHHHHHHHHHHHHH. conspiracy
GTaudiophile
Lifer
- Oct 24, 2000
- 29,767
- 33
- 81
Originally posted by: XBoxLPU
Umm Apple test shows 1024X768, tom's is 640X480
Someone just got pwned.
If the benchmark utilized vectorized code then yes it's highly possible, icc is that much better. GCC doesn't really optimize for SSE2 as well as ICC.
Yes. I meant 402. And that's with a slower VPU.
It changes nothing - Apple results are very misleading, showing 40% delta between their P4 results, and independent benchmarks. Given Apple sleazy marketing in the past, I am not surprised.
p.s. According to the wp, they used GCC, as I expected. Ah, good old Apple
Leon
It changes nothing - Apple results are very misleading, showing 40% delta between their P4 results, and independent benchmarks. Given Apple sleazy marketing in the past, I am not surprised.
p.s. According to the wp, they used GCC, as I expected. Ah, good old Apple
Leon
Numbers are just numbers IMO and people get wrapped up in benchmarks way too much.
I still would love to own a G5
I still would love to own a G5
Well, I'd say wait and see how 10.3 turns out and in the meantime wait for some non-Apple benchmarks. IIRC Carmack pointed out AltiVec does not help that much with Q3A and I doubt people who'd get a G5 (or a dual Xeon, for that matter) would play Q3A on it. Where I think the G5 will kick a$$ is in application that need, say, 8GB or RAM and a 6.4 GB/sec bandwidth. Q3A does not fall under that category.
On another note, are the Power4 SPEC numbers single or dual core? If they are single one can make a pretty nice conclusion on how IBM's compiler (XFsomething?) would do with a 970 (non-AltiVec-optimised).
On another note, are the Power4 SPEC numbers single or dual core? If they are single one can make a pretty nice conclusion on how IBM's compiler (XFsomething?) would do with a 970 (non-AltiVec-optimised).
Random Variable
Lifer
- Aug 10, 2001
- 10,420
- 2
- 0
Eug
Lifer
- Mar 11, 2000
- 24,167
- 1,812
- 126
It doesn't optimize well for Altivec either does it? (Asking.)
Cheaper, and Apple has this habit of under-spec-ing the low end model presumably to make people spend more money to get the mid-end. On the downside, it means inferior specs, but on the upside it's makes it reachable for those with a very limited budget.
Generally, if you get a Mac, get the mid-end or up. With current prices, the mid-end is more affordable now fortunately.
I still have the picture in my mind of the woman buying the G4 Power Mac on Saturday. I soooooooooooooooooooooo wanted to tell her to wait until today if she could but I kept my mouth shut. I was at the Apple Store in Atlanta, picking up a new iPod. It ROCKS by the way.
Originally posted by: Leon
Yes. I meant 402. And that's with a slower VPU.
It changes nothing - Apple results are very misleading, showing 40% delta between their P4 results, and independent benchmarks. Given Apple sleazy marketing in the past, I am not surprised.
p.s. According to the wp, they used GCC, as I expected. Ah, good old Apple
Leon
The thing about gcc is that it's really not optimized that well for PowerPC. I really don't think it was a marketing decision to use gcc to make the P4 look bad. The new Apple has been advocating standards for a while, and part of that is using standardized tools like GCC. The "official" compiler of OS X is GCC, so for us OS X developers, it makes a lot of sense to use GCC benchmarks.
I'm trying to look past Steve's RDF as I am a fairly recent Apple newbie having purchased an iBook. But I think it's entirely reasonable to say that these benchmarks released by Apple are more true-to-life than any others of recent memory. There was no need for releasing on Photoshop benchmarks using highly concocted and biased routines to make the Mac look better. I'm sure an indepedent review will show that Apple is being a little imprecise trying to gain a few percentage points in a benchmark here and there. But overeall, I think the outcomes are pretty accurate.
$2000 with no monitor is for those with a very limited budget? Where do you live? For $2000 I can get a Dell with 800MHz P4 2.8 w/ 1GB RAM and a 9800Pro and Audigy 2 and 120GB drive.
If you are on a very limited budget Apple, is no the way to go.
Random Variable
Lifer
- Aug 10, 2001
- 10,420
- 2
- 0
http://store.apple.com/1-800-MY-APPLE/WebObjects/AppleStore.woa/72201/wo/rv5Pd7v1SaeO2HwUmV61MryiVDi/7.0.7.1.0.5.1
Fully decked out G5... $13,730.90
Ay caramba!
Fully decked out G5... $13,730.90
Ay caramba!
GTaudiophile
Lifer
- Oct 24, 2000
- 29,767
- 33
- 81
Yeah, and like 50% of that is in the monitors. Not that I wouldn't cream at the site of two 23" Apply Displays
Eug
Lifer
- Mar 11, 2000
- 24,167
- 1,812
- 126
Errr.... Of course it's gonna be that much, considering how you spec'd it:
8 GB RAM
DUAL 23" Cinema Displays (at $2000 each)
500 GB of hard drive space (two drives)
Wireless 802.11g
Wireless base station
Fibre channel PCI card (which is $500 alone)
30 GB iPod
BlueTooth module
.Mac internet acct.
High-end 5.1 speakers.
Extra 3 year warranty
Eug
Lifer
- Mar 11, 2000
- 24,167
- 1,812
- 126
Oops. It seems I'm way off base on this one. There is no such thing as an 800 MHz FSB Xeon. Unlike the P4, all are 533 or slower. 800 MHz Xeons won't be out until 2004, and that's only with the 90 nm process.Originally posted by: Eug
Hmmm... I'm seeing dual Xeon in more than one place, but it's 533 MHz Xeons.
So not a direct comparison with 800 MHz and HT, but still, it's DAAAAMMMMNNN fast.
And IBM has announced a 3 GHz G5 within 12 months!
http://intel.com/products/server/processors/server/xeon/index.htm?iid=ipp_srvr_proc+high1_xeon&
From CNN.com: ""Apple could have a winner on its hands, especially when it introduces a Windows version later this year," Wolf wrote."
BFG10K
Lifer
- Aug 14, 2000
- 22,709
- 3,007
- 126
Yikes, we won't see them until September 1st and by that time Intel will probably have 3.5 GHz P5s out. Also Apple's PC SPEC and Quake 3 benchmarks are far lower than real scores, plus HT was disabled in the tests.
Also using benchmarks from the likes of Tom a 2.4 GHz P4 beats Apple's own Quake 3 benchmarks from their dual-2 GHz systems. And I can tell you now that a 2.4 GHz P4 system is not even close to costing $3000 like Apple's machines do. So much for Apple's Quake 3 "graphs".
Also using benchmarks from the likes of Tom a 2.4 GHz P4 beats Apple's own Quake 3 benchmarks from their dual-2 GHz systems. And I can tell you now that a 2.4 GHz P4 system is not even close to costing $3000 like Apple's machines do. So much for Apple's Quake 3 "graphs".
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter