| ActiveMac | Anonymous | Create a User | Reviews | News | Forums | Advertise |
|
|
|
User Controls |
|
New User |
|
Login |
|
Edit/View My Profile |
|
|
|
|
|
|
|
ActiveMac |
|
Articles |
|
Forums |
|
Links |
|
News |
|
News Search |
|
Reviews |
|
|
|
|
|
|
|
News Centers |
|
Windows/Microsoft |
|
DVD |
|
ActiveHardware |
|
Xbox |
|
MaINTosh |
|
News Search |
|
|
|
|
|
|
|
ANet Chats |
|
The Lobby |
|
Special Events Room |
|
Developer's Lounge |
|
XBox Chat |
|
|
|
|
|
|
|
FAQ's |
|
Windows 98/98 SE |
|
Windows 2000 |
|
Windows Me |
|
Windows "Whistler" XP |
|
Windows CE |
|
Internet Explorer 6 |
|
Internet Explorer 5 |
|
Xbox |
|
DirectX |
|
DVD's |
|
|
|
|
|
|
|
TopTechTips |
|
Registry Tips |
|
Windows 95/98 |
|
Windows 2000 |
|
Internet Explorer 4 |
|
Internet Explorer 5 |
|
Windows NT Tips |
|
Program Tips |
|
Easter Eggs |
|
Hardware |
|
DVD |
|
|
|
|
|
|
|
Latest Reviews |
|
Applications |
|
Microsoft Windows XP Professional |
|
Norton SystemWorks 2002 |
|
|
|
Hardware |
|
Intel Personal Audio Player 3000 |
|
Microsoft Wireless IntelliMouse Explorer |
|
|
|
|
|
|
|
Site News/Info |
|
About This Site |
|
Affiliates |
|
ANet Forums |
|
Contact Us |
|
Default Home Page |
|
Link To Us |
|
Links |
|
Member Pages |
|
Site Search |
|
Awards |
|
|
|
|
|
|
|
Credits |
 |
Apple intros Panther, iSight, iChat AV and the new G5's | |
|
||
| Time: 14:20 EST/19:20 GMT | News Source: TopTechTips | Posted By: Brian Kvalheim | ||
|
"We have so much good stuff for you today..." Safari SDK (Dev Toolkit) today Safari Gold! Today - Will be released in 2/3 hours time. Jaguar is dead Panther has a brand new Finder, old one was "too computer centric" Fast Searching Action button added - which shows u what you can do with a document Labels are back Dynamic Network Browser - search for specific servers Enhanced iDisk significantly for Panther Finder Windows all brushed metal. | ||
|
Write Comment
Return to News |
Displaying 26 through 50 of 361 Prev | Last | Next |
The time now is
10:59:41 AM
ET.
Any comment problems? E-mail us |
| #26 By 3339 (66.219.95.6) at 6/23/2003 6:48:01 PM |
|
bluvg, our Optiplexes have the door that swings out the side, but if you use them as towers, you have a release button UNDER the box! Under the box? Two buttons you have to release, not one, and one is on the bottom... And then it opens like a book standing up instead of dropping down, so you have to drag the side along the surface of whatever the box is on. And then you have components on both sides, but the opened access is in the back... So you are still talking about picking up, moving the machine around, making space around the work area....
I don't know how that's easier. |
| #27 By 3 (62.253.128.4) at 6/23/2003 6:51:13 PM |
| #26 - its an official shot from Apple PR - it just went up on the Apple PR site too, its the same, guess it is an Apple mistake. |
| #28 By 2459 (69.22.78.116) at 6/23/2003 7:06:18 PM |
| Apple's trying to outdo the case modding crowd. 9 fans. I think that's about 4 more than my full tower. |
| #29 By 2459 (69.22.78.116) at 6/23/2003 7:48:52 PM |
|
Here's the issue though:
If Apple's SPEC tests aren't accurate. How can one tell if their other benchmarks don't suffer similar inaccuracies? In regards to SPEC, the system I checked was the same system Apple used in their benchmarks (Precision 650 3.06GHz Xeon, IIRC). Also keep in mind that some AthlonXP (and I'm betting Opteron) also beat out the G5 in SPEC. Besides, if it isn't already, that Dell system will likely be cheaper by the G5's availability due to the cost of hardware and the CPU lowering at a faster rate as usual. Again, Opteron would provide a better comparison in any case because it's cheaper and more performant than Xeon last I checked. Found it: http://www.amd.com/opteronperformance or search http://www.spec.org which is where I got the base int and fp Opteron numbers (AMD only showed peak). SPECint G5 2GHz - 800 Opteron 144 (1.8 GHz) - 1095 SPECfp G5 2GHz - 840 Opteron 144 - 1122 SPECint_rate G5 2GHz - 17.2 Opteron 244 (1.8 GHz) - 26.8 SPECfp_rate G5 2GHz - 15.7 Opteron 244 - 26.7 Even the Opteron 240 which runs at 1.4 GHz had faster SPEC marks. This post was edited by n4cer on Monday, June 23, 2003 at 19:56. |
| #30 By 3339 (65.198.47.10) at 6/23/2003 8:05:40 PM |
|
Enforcer, they aren't faked. They are SPEC scores with a different OS and compiler. There is no discrepency as there is no equivalence.
There are no Opterons shipping... Why would Apple compare themselves to something vendors won't have for a month? The dual Xeon is going to be cheaper than the PowerMac? Don't think so. And, again, are you saying that I/O, memory bandwith, network access, vector calculations, multithreading of apps and other software customizations are irrelevent in comparison to fp and integer? Just curious. |
| #31 By 2459 (69.22.78.116) at 6/23/2003 8:22:35 PM |
|
I didn't say they were faked, just innaccurate as far as saying the G5 is the fastest PC.
Opterons have been shipping for some time. I think you're thinking of the Athlon64. I/O, bandwidth, etc., aren't irrelevant, but the problem with the other comparisons is that where one fails in showing Apple is the fastest, so can the others. Especially by shipping time. The main point I'm trying to make here is that Apple's claims of having the fastest PC are questionable. Sure, the G5 is a big step up from the G4 and it appears to have the architucture to compete with Intel and AMD, but based on the differences in SPEC (which Apple is using to make the claim of "fastest"), I think there is enough evidence to draw reasonable doubt to that claim, especially because Apple has never been too open with their benches, and there have been many independents to dispute them in the past with reproducible results Again, don't take this as me saying the G5 isn't good or competitive. I'm simply saying there is reasonable doubt as to Apple's "fastest" claim. |
| #32 By 2459 (69.22.78.116) at 6/23/2003 8:28:06 PM |
| Note: the "open benches" comment refers to the benches other than SPEC, such as Photoshop and other tests where there is no published method or script to reproduce results or full disclosure of system specs. |
| #33 By 3339 (66.219.95.6) at 6/23/2003 8:52:29 PM |
|
"Opterons have been shipping for some time. I think you're thinking of the Athlon64. "
You'd know better than me... I'll leave it to you. But show me a box, a complete system, from Dell or Hp or whoever with a Opteron mobo. Mobos that aren't reliable enough to actually result in a shipping product don't impress me. I haven't seen a Dell or HP Opteron system... I can't find one quickly at dell. In fact, it looks like IBM may be the first, and they are only going to preview it as well later this week. But, as I said, let me know, I'd appreciate it. "Theoretical" systems that don't have a complete architecture are a joke. SPEC measures the performance of a hardware configuration (not just the proc but the controller, buses, etc...), the compiler, and the OS. "The main point I'm trying to make here is that Apple's claims of having the fastest PC are questionable." When they are no longer valid they won't make the claim. You'll still hold it against them though, acting as if they continue to make the claim. All SPEC tests require reading the fine print: what OS, what are the interconnects, what controller is used, what compiler, etc. If you use SPEC marks any other way then taking note of these details, you aren't using SPEC correctly and you are deceiving yourself. So I don't see how failing to read the fine print in this case equals "deception." "Note: the "open benches" comment refers to the benches other than SPEC, such as Photoshop and other tests where there is no published method or script to reproduce results or full disclosure of system specs." You mean the ones that Adobe stands by. The specs are available. Show me a single independent benchmark that invalidates the PS bakeoffs. This post was edited by sodajerk on Monday, June 23, 2003 at 21:43. |
| #34 By 3339 (66.219.95.6) at 6/23/2003 9:05:45 PM |
|
"Again, don't take this as me saying the G5 isn't good or competitive." I'm sure you are not.
But let's just say what we do have: A system which is comparable (slightly better or slightly worse or more likely a bit of each at different tasks) to the best hardware configs available today in the PC world. A system which will continue to be competitive with the future of the PC world without the limitations or infighting present in the whole Itanium/Opteron/Athlon64 battle. (Purely from a HW computational perspective.) A system which is dramatically cheaper than the most appropriate comparison available now (dual Xeons or worse, proprietary Unix stations). (Admittedly, I'm primarily concerning myself with the high-end of Apple's offering rather than a low end Mac vs. a cheap, home-built PC... But (once again) the PM is a machine for businesses and pros.) A system with Dual 1Ghz FSBs, PCI-X, Serial ATA with dual independant channels, AGP 8X Pro(for more juice), FireWire 800 and 400 and USB 2, Hypertransport interconnects, 8GB Dual Channel DDR 400 ram, digital optical audio in/out, gigabit ethernet, internal bluetooth and 802.11g (BTO). [Let's note that there isn't a single PC mobo available with the dual 1 GHz fsb, PCI-X, or S-ATA w/indepedent channels. Nevermind all of these features.] A system with an ugly box? You decide. This post was edited by sodajerk on Monday, June 23, 2003 at 21:10. |
| #35 By 2459 (69.22.78.116) at 6/23/2003 10:40:51 PM |
|
You're wrong about no PC motherboards having PCI-X, they're mainly workstation/server class boards, though PCI-Express will take over for most systems in the next few months and will likely be more popular. Dual independent channel SATA is probably available as well, though I'm not sure. It really doesn't matter as long as the bus provides enough bandwidth so as not to become saturated.
As you implied earlier, what matters is the system as a whole. Apple is competitive here, but I still doubt their claims of being fastest. As far as "theoretical systems" go, the G5 is such a system until Apple ships then in the next few months. |
| #36 By 3339 (65.198.47.10) at 6/23/2003 11:08:16 PM |
|
Well, enforcer, there are plenty of people who are getting them: Genetech, Pixar, others... You'll note they are even soliciting customers to ask for a service rep call... If you have the right business or big enough order, you could get one now. Others are ordering them today. You can't even order IBM's Opteron workstation.
PCI-X is hardly available... come on... If some obscure manufacturer is building it into mobos, you can point that out to me... But I'm still waiting on you to point out a shipping Opteron system. No, S-ATA isn't available, nevermind indepedent channels... Come on, parts that haven't or weren't supposed to get out of the lab don't count. ...And somehow you failed to point out that box that has a 1 GHz fsb? and, you know, the other one with dual independent 1 GHz fsbs? And again, is there a system that has Dual 1Ghz FSBs, PCI-X (at 133 and 100 MHz), Serial ATA with dual independant channels, AGP 8X Pro(for more juice), FireWire 800 and 400 and USB 2, Hypertransport interconnects, 8GB Dual Channel DDR 400 ram, digital optical audio in/out, gigabit ethernet, internal bluetooth and 802.11g (BTO) all on ONE mobo? And in terms of raw performance, is this competitive with dual Xeons? How many people do you know running dual Xeons? And how much would that cost them? Nevermind Opterons or Athlon64s? How many of your friends have bought that phantom Opteron system? (By the way, since when is AMD saying Opteron is for desktop use?) And your not-open "open benches"? Here you go: http://a192.g.akamai.net/7/192/51/ebb34a6c95daa5/www.apple.com/powermac/pdf/PowerMacG5_Perf_WP_062303.pdf Go to the last page. They publish this every time they have executed these tests. So where's the "indepedent test" where the same tests reveals contrary results? Please, enforcer, you haven't backed up one thing today. Most of your posts were ill-informed, weak-kneed whining. This post was edited by sodajerk on Monday, June 23, 2003 at 23:15. |
| #37 By 2459 (69.22.78.116) at 6/23/2003 11:33:23 PM |
|
Oh, please. How many of your friends have bought a phantom G5? It's not like you can take one home right now.
SPEC tests on spec.org refute Apple's claims of fastest PC. Other than that, you'll have to wait until someone actually can buy a G5 and run their own tests. Opterons are the only way to get dual CPUs with the AMD64 architecture like AthlonMP is with AMD32. You can build a dual Opteron system at a price competitive with the G5. Of course, according to SPEC, you only need one Opteron to be competitive. Purchase an Opteron Workstation (complete with nforce3 and GeForce FX) http://www.xicomputer.com/products/mtowerop.asp PCI-X motherboards/systems http://www.google.com/search?hl=en&ie=UTF-8&oe=UTF-8&q=pci-x+motherboard SATA motherboards http://www.google.com/search?hl=en&lr=&ie=UTF-8&oe=UTF-8&q=sata+motherboard&btnG=Google+Search About the 1GHz FSB -- still no one has told me whether Apple's memory can actually take advantage of this, and, IIRC, memory bandwidth is still only 6.4 GB/s. As far as I can tell, the GHz FSB is just from CPU to system controller. What's the GHz bus' actual throughput? |
| #38 By 135 (208.50.204.91) at 6/24/2003 1:50:46 AM |
| This is wonderful! The one thing preventing widescale Linux adoption is a case that looks like a cheese grater! |
| #39 By 7754 (24.245.76.135) at 6/24/2003 3:46:28 AM |
|
Sodajerk,
As I said, show me a workstation that needs more than 500GB of local storage and I'll show you an individual or company that doesn't know sh!t about networking. (Are you really suggesting you want/need more HD capacity INSIDE your workstation?) That is exactly what I am suggesting. When you are working with huge video files and the like, even gigabit ethernet (and the associated TCP/IP and ethernet overhead) is no comparison to working on the files locally. Some programs don't work well with large files over the network, also; for example, at the last place I worked, our Quark users got sick of dealing with corruption and other issues and would just download the entire project files to work on them locally, uploading them when finished. The network is a great way to collaborate on files and to centralize them for backup, but performance-wise it is a distant second place to working on files locally, even with proprietary networking hardware. Networking aside, I can see a desire to insert other devices into the machine, and the only option will be to set them up externally. Hey, it's not a big problem, but for some it's a drawback. As far as the grate design, my concern is that it's right on the front of the case. Others use similar holes as well, but I haven't seen any that put them right in the front. It's totally a minor issue, though... I'm not trying to nit-pick, it's just a comment. That's somewhat interesting that only the SMP systems have the option for the 8 GB of RAM--perhaps a current chipset limit? I guess that people interested in that much RAM would likely be going for a dual or more anyhow, though. As for the benchmarks, I will wait until others are able to independently test the machines. As n4cer pointed out (as well as many industry analysts), the 970 will still be behind AMD and Intel in many benchmarks by the time it's shipping (and it's already behind the Opteron). After seeing what has happened in the past with Apple announcing benchmarks, I will read their claims about the G5 with both interest and suspicion. I will also be curious to see the full picture of how all the current software and peripherals fare with the G5 and G5-ready OS. Hopefully it will be a seamless transition for them. Also, as n4cer pointed out, there are a few things that do not add up--for example, the 1 GHz FSB. Yet, the RAM is only 400 DDR, and the HyperTransport interface runs at 800 MHz--another Porsche in a parking lot problem? A performance increase for sure, but some of the numbers need to be evaluated in the light of the achievable vs. the theoretical. This post was edited by bluvg on Tuesday, June 24, 2003 at 03:55. |
| #40 By 7754 (24.245.76.135) at 6/24/2003 3:46:56 AM |
|
With the Optiplexes, ours all have the release button on the front of the case. I haven’t seen the ones with the button on the bottom. I just know that when it comes to replacing anything in the machine, it’ll take less than 5 minutes… even a full motherboard swap. The cases are very solid as well. If they could reduce the clutter as well as Apple has with the G5, that’s cool, but I’ll be happy as long as they stay so easy to fix.
As a side note, I saw a shipping Opteron system at a Windows 2003 Server launch. Don’t remember the name of the brand, though, but I guess it doesn’t matter as n4cer has posted a link to a manufacturer. When they are no longer valid they won't make the claim. You'll still hold it against them though, acting as if they continue to make the claim. True, because the Apple faithful will still believe the claim and propagate it. The myth that the G4 blows away any Intel system still exists… the PC manufacturers don’t release new products with great fanfare saying, “…eats any G4 for lunch” yada yada yada, so the myth continues. I know too many people that should know better and still believe this. You’re pointing out a lot of things, though, that still compare this non-shipping product with shipping products (and yes, there are non-production machines available in the Intel/AMD world as well that can be obtained if you are “special,” so I don’t think the statement that you can obtain these from Apple is all that useful): The benchmarks aren't that crazy... They use parts that Dell actually ships, not ones they plan to introduce in a month or a couple of weeks. Yes—exactly—since they are comparing Dell shipping parts with Apple non-shipping parts, the benchmarks are a bit crazy. If you want to use them as a baseline for performance rather than the current G4 (which would have made more sense… and would have been much more interesting to see, as it would point out just how pokey the current Macs are… very interesting indeed that they didn’t do that… the only comparison I saw between the G5 and the G4 was that the G5 system is twice as quiet! Wise move, Apple, wise move…), that’s fine, but they should be considered just that—a baseline—rather than a competitive comparison. I think we may just end up having to wait until August to see what the situation is. No matter what, it’s a great leap forward for Apple, and finally an answer to their G4’s lackluster performance issues. It would have been nice if Motorola had been able to keep up so that Apple had competition within their own platform as the PC side has with Intel and AMD, though… I guess they will continue to have only indirect competition (of course, IBM is a better partner choice than Motorola, I think… the 1984 commercial irony aside…). I’m looking forward to seeing what kind of pressure this puts on Intel and AMD, though, as well as on 64bit computing for desktops. This post was edited by bluvg on Tuesday, June 24, 2003 at 04:01. |
| #41 By 7754 (24.245.76.135) at 6/24/2003 3:50:22 AM |
|
BTW, sodajerk, you may want to retract your comments (and hard drive keepsake) about Steve Ballmer/Microsoft PR from earlier: :)
http://news.com.com/2100-1043_3-1019181.html This post was edited by bluvg on Tuesday, June 24, 2003 at 04:00. |
| #42 By 7754 (216.160.8.41) at 6/24/2003 12:34:58 PM |
|
Thanks, Parker... here's another (linked from your post):
http://www.amdzone.com/articleview.cfm?articleid=1296 |
| #43 By 2459 (69.22.78.116) at 6/24/2003 8:22:03 PM |
|
Thanks Parker. This is some pretty funny (and shady) stuff.
Why does Apple keep doing this kind of stuff (retorical)? |
| #44 By 2459 (69.22.78.116) at 6/24/2003 10:12:57 PM |
|
A couple more links:
Apple Hardware VP Defends Benchmarks http://apple.slashdot.org/article.pl?sid=03/06/24/2154256&mode=thread&tid=126&tid=181 Apple Benchmarketing Raises Questions http://www.extremetech.com/article2/0,3973,1136018,00.asp |
|
Write Comment
Return to News |
Displaying 26 through 50 of 361 Prev | Last | Next |
The time now is
10:59:41 AM
ET.
Any comment problems? E-mail us |
