*****CONFIRMED**** PIV slower in Spec tests with HT enabled. Apple not lying about that.
- Thread starter EdipisReks
- Start date
Is there a software application that I can download to toggle HT on and off on a remote server? Obviously I cannot access the BIOS on a remote machine...
-DAK-
Well, personally, i don't find this really surprising--I've noticed it quite a bit on the workstations at school. Its not an enormous hit..... but its there (it shows up during my runs). I'm not sure if its "reserving" processor capacity or what, though... because its certainly easier to open up a new program with HT enabled.
Oh well... for most multi-tasking users, I still think its a good thing. If you're doing something so complicated that the extra 2% makes a REAL difference, just turn it off in the BIOS when you're doing it.
Oh well... for most multi-tasking users, I still think its a good thing. If you're doing something so complicated that the extra 2% makes a REAL difference, just turn it off in the BIOS when you're doing it.
I would have to agree that the new macs are very nice. But they ARENT EVEN CLOSE to a value. A PC with the kind of budget it takes to buy the new mac would be A LOT faster... You could easily afford RAID5, 3ghz, 2GB of ram, I875, and prolly a real stereo system + Plazma TV.
Eug
Lifer
- Mar 11, 2000
- 24,048
- 1,679
- 126
Haha. Own3d.
Dell themselves say: "Incorporated into Intel Xeon processors, Hyper-Threading technology can provide great benefits to server applications. Unfortunately, it can also degrade system performance in certain scenarios such as those simulated by the compute-intensive SPEC CPU2000 and Linpack benchmarks."
Originally posted by: Eug
Haha. Own3d.
Dell themselves say: "Incorporated into Intel Xeon processors, Hyper-Threading technology can provide great benefits to server applications. Unfortunately, it can also degrade system performance in certain scenarios such as those simulated by the compute-intensive SPEC CPU2000 and Linpack benchmarks."
Yeah. I've always thought that, except for a few best-case scenarios, hyper-threading lowers benchmark results but increases "interactive latency" for lack of a better term. The system should feel more responsive but it won't necessarily fly through benchmarks faster.
Did ANYONE notice that the article was written almost ONE YEAR AGO? That's when HT wasn't fully finished, even Anand wasn't excited about it then. However, today, it's a much different story.
Eug
Lifer
- Mar 11, 2000
- 24,048
- 1,679
- 126
How was it "not finished"?Did ANYONE notice that the article was written almost ONE YEAR AGO? That's when HT wasn't fully finished, even Anand wasn't excited about it then. However, today, it's a much different story.
These are Xeons with HT, which predated P4s with HT by quite some time.Oh and the authors of the article are:
Humayun Khalid, Ph.D. (humayun_khalid@dell.com) is a performance architect for the Dell System Performance and Analysis Lab. His responsibilities include performance evaluation, analysis, and projections for Dell servers. He is also responsible for design support, performance modeling, marketing support, competitive analysis, and chipset-architecture-technology (CAT) evaluation. Humayun has a Ph.D. in Electrical Engineering from the City University of New York.
Ramesh Radhakrishnan, Ph.D. (ramesh_radhakrishnan@dell.com) is a design engineer consultant with the Dell System Performance and Analysis Lab. His responsibilities include performance analysis of Dell servers and characterization of enterprise-level benchmarks. Ramesh received a Ph.D. in Computer Engineering from the University of Texas at Austin.
Nicholas Wakou (nicholas_wakou@dell.com) is a systems engineer consultant with the Dell System Performance and Analysis Lab. His responsibilities include generating TPC-C® benchmark performance and price/performance numbers. He received his M.E. in Electrical Engineering from Oklahoma State University.
I agree that Anand is a very smart guy, but quite frankly I don't see how Anand's opinion would supercede these guys' scientific findings. Furthermore, I don't remember reading anywhere that Anand directly disagrees with Dell's conclusion about SPEC anyway.
Apple makes the whole package... including the hardware, the OS, and even much of the software. Because of this, it's no surprise that Apple stuff "just works". Sure, it ain't perfect, but it's more likely you'll have less technical issues with a mainstream Apple setup than a mainstream Wintel setup. (I have both by the way.) Wintel setups can be tweaked by the end user, but Apple removes much of the requirement for this tweaking, because they tweak their OS and software to their own hardware (because they make it all).
Furthermore, Mac OS X is much nicer IMO. As for stability, I don't really know because both my machines are pretty stable. I can honestly say that almost never reboot my OS X laptop, unless I'm installing OS upgrades or something. My WinXP box is pretty damn stable too, but I'd give the edge to my OS X laptop, despite the fact that I've spent much more time tweaking my XP box, and the laptop gets used much more .
It's to make money. You really think Apple cares about people? Don't flatter yourself. They only care about fattening their wallets. Defend them all you want, but atleast know who you are REALLY defending.
From what I have read, earlier Xeons HT was not as good as what PIVs HT are now. I am not for sure, but I do think intel updated HT in newer Xeons also
It wasn't finished, as someone who worked on Willamette told me once. There were quite a few changes introduced in Northwood and its later stepping (the "C" stepping I believe, for the 3.06 GHz model) that dramatically improved the arbitration of resources under SMT, especially for the memory hierarchy (this has been discussed by Intel at previous IDFs). HT was enabled under the Willamette-based Xeons earlier than the Northwood P4s because of increased demand.Originally posted by: Eug
How was it "not finished"?Did ANYONE notice that the article was written almost ONE YEAR AGO? That's when HT wasn't fully finished, even Anand wasn't excited about it then. However, today, it's a much different story.
Looking at the previous articles in the Dell issue (link at the top of the linked page), it is implied that they indeed tested the Willamette-based Xeon a year ago. I would not be surprised if the results would be significantly higher if they were reperformed today.
Originally posted by: Sohcan
It wasn't finished, as someone who worked on Willamette told me once. There were quite a few changes introduced in Northwood and its later stepping (the "C" stepping I believe, for the 3.06 GHz model) that dramatically improved the arbitration of resources under SMT, especially for the memory hierarchy (this has been discussed by Intel at previous IDFs). HT was enabled under the Willamette-based Xeons earlier than the Northwood P4s because of increased demand.Originally posted by: Eug
How was it "not finished"?Did ANYONE notice that the article was written almost ONE YEAR AGO? That's when HT wasn't fully finished, even Anand wasn't excited about it then. However, today, it's a much different story.
Looking at the previous articles in the Dell issue (link at the top of the linked page), it is implied that they indeed tested the Willamette-based Xeon a year ago. I would not be surprised if the results would be significantly higher if they were reperformed today.
how much cache does the Williamette Xeon's have? that might affect performance. i'm not in the server side of chips, so i don't know
sohcan, are you an Intel employee? just got interested
I don't own a Mac but my friend is a devout Ti Powerbook user since he got one last year and replaced with desktop Athlon system with it. He honestly does far far more tweaking than I do on my XP machine (I spent about 2 hours including installation and patching to get my OS the way I wanted it and 100% stable). I think it is because of the large amount of unix programs he uses and tweaking the source before compiling. I still think a unix based system requires more tweaking depending on the software you want to use versus XP. This doesn't apple to native apple software of course.
Personally I think Apple sells an image or style, non-mainstream and creative/independent. You always see those powerbook users in their black turtlenecks with black rimmed glasses sipping coffee at the Starbucks The only practical benefit I see of their solutions is in graphics/sound production and out of the box useability.
Personally I think Apple sells an image or style, non-mainstream and creative/independent. You always see those powerbook users in their black turtlenecks with black rimmed glasses sipping coffee at the Starbucks
The only practical benefit I see of their solutions is in graphics/sound production and out of the box useability.
Eug
Lifer
- Mar 11, 2000
- 24,048
- 1,679
- 126
OK, that is a fair point. I would like to see someone do the SPEC testing with the current Xeon (besides Apple of course) with HT on and off. This is not to deny HT can have very significant real-world benefits, because we all know that it can. It is just to see how the Xeon +/- HT would do in the SPEC benchmark, since this is the issue at hand.Originally posted by: Sohcan
It wasn't finished, as someone who worked on Willamette told me once. There were quite a few changes introduced in Northwood and its later stepping (the "C" stepping I believe, for the 3.06 GHz model) that dramatically improved the arbitration of resources under SMT, especially for the memory hierarchy (this has been discussed by Intel at previous IDFs). HT was enabled under the Willamette-based Xeons earlier than the Northwood P4s because of increased demand.Originally posted by: Eug
How was it "not finished"?Did ANYONE notice that the article was written almost ONE YEAR AGO? That's when HT wasn't fully finished, even Anand wasn't excited about it then. However, today, it's a much different story.
Looking at the previous articles in the Dell issue (link at the top of the linked page), it is implied that they indeed tested the Willamette-based Xeon a year ago. I would not be surprised if the results would be significantly higher if they were reperformed today.
For what it's worth, Apple claims it chose to turn off HT on the current CPUs specifically because it decreased performance (even with the current Xeon) in the SPEC benchmarks. They also chose Linux over Windows, because Linux was faster. They also went on to say they'd be willing to repeat the benches with Windows and with HT turned on.
They DID admit that the Intel compiler would be faster, but go on to say it is because it is heavily optimized for the Xeon, whereas GCC isn't all that greatly optimized for either CPU (albeit probably better for the Xeon). I wonder how good the Code Warrior compiler is on the Mac side by the way. In the benches Apple itself showed, compile times with Code Warrior simply blew away GCC. I don't know how this translates into performance of the compiled app however.
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 Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter