Tom Hardware Stress Test

Quote:

---

Around this same time last holiday season, we started a project that ended up setting a world record for processor speeds. Originally, we had planned to content ourselves with the ambitious enough 5 GHz Project: CPU Cooling With Liquid Nitrogen. In the end, we achieved a maximum speed of 5.25 GHz at -190°C. Chipmakers aren't likely to start offering production models capable of such speeds until 2006 at the earliest!

So what about this year? No new speed record? After discussing the issue ad nauseam, we finally decided to shelve record-breaking in favor of revisiting the systems currently in use in real life - where stability and product quality are the primary factors in making a buying decision. This is especially true of the business environment. Here the critical decisive factor for any investment is the TCO (total cost of ownership) over a period of time. So we thought the obvious thing to do would be to subject the latest PC system components to a demanding stress test under real-life conditions.

---

Source : http://www.tomshardware.com/cpu/20041217/index.html
Stress Test: http://www.tomshardware.com/stresstest/index.html

Intel as a reboot! AMD none!

Amd owns Intel products ANY day.

Acidloop, sorry to burst your bubble, but AMD still loses to Intel in media encoding benchmarks. A longer pipeline and hyperthreading really do tend to benefit it beyond what AMD can offer at the moment.

The Intel machine's reboot was purportedly caused by a faulty power supply.

Reposting what I've already posted elsewhere on this subject:
On the power supply front, IMO an infinitely more intelligent approach would have been to use high quality redundant server power supplies. This would have two effects:
1. If one of the redundant supplies died, it would switch to the hot backup with no downtime.
2. Server power supplies tend to be far better regulated, and therefore you pretty well eliminate the power supply as a potential instability.

There are other considerations to be made. As they are running a single system of each type, any failures that do occur may be the result of abnormalities. This has also been mentioned before (elsewhere), but perhaps a multiple system test would be a better indicator of overall stability.

Additionally, it would appear that somehow the Intel system's power failure adversely affected the AMD system for a short duration. It appears to me that the "week overview" graphs show a drop in fan RPM on the AMD system's Northbridge fan that looks like it coincides directly with the Intel system's power outage. This leads me to suggest that a future test should have both systems hooked up to uninterruptible power supplies, preferably on isolated circuits.

Moving on, I also find the use of a graphic-intensive benchmark rather dubious. The actual stress this puts on the processor is unknown. I find myself asking why a processor platform stability test focuses on graphics.

The test wasn't exactly created using the scientific method, however I do appreciate that this test is better than no test, and there will likely always be people with issues in the methodology.

Quote:

---

Originally posted here by chsh
Acidloop, sorry to burst your bubble, but AMD still loses to Intel in media encoding benchmarks. A longer pipeline and hyperthreading really do tend to benefit it beyond what AMD can offer at the moment.

---

That is not true anymore. AMD Socket 939 with the latest core is as good that any Intel in media encoding benchmarks now. You can read it here http://www.materiel.be/cpu/2004cpuv3/page1.php

It's in French but I'm sure you can understand the graphic.

I read all the artical SDK and I can say only: AMD RULES!!!!

Quote:

---

Originally posted here by SDK
That is not true anymore. AMD Socket 939 with the latest core is as good that any Intel in media encoding benchmarks now. You can read it here http://www.materiel.be/cpu/2004cpuv3/page1.php
It's in French but I'm sure you can understand the graphic.

---

As good as in one video-related benchmark on that site, and that deals only with compression to specific codecs. There are no benchmarks displaying Video Editing software, which generally gets far better boosts from HyperThreading than (comparatively) simple math-oriented compression.

Example: http://www.anandtech.com/cpuchipsets...spx?i=2275&p=8 (bottom of the page).

None of what I said is a shot at AMD, it was merely a shot at product cheerleading. They have begun to dominate performance in pretty much all other aspects of x86 computing (check Anandtech's DB and Web Server comparisons for a giggle). Consumers really only benefit from competition between the two companies.

Just FYI, I have owned every major core revision of AMD chips going back to the K6-2 up to and including the Semprons, so it's not like I'm bashing them.

When it comes to heavy processing, Intel will take AMD in a clean sweep.

I have both a P4 3.2 and a AMD 64 3200. When it comes to video editing or playing the newest games I can tell that there is a difference between the AMD and teh Intel chipset. Games and Video apps just seem to run and play smoother with a Intel chip in the board.

Quote:

---

Originally posted here by cheyenne1212
When it comes to heavy processing, Intel will take AMD in a clean sweep.

I have both a P4 3.2 and a AMD 64 3200. When it comes to video editing or playing the newest games I can tell that there is a difference between the AMD and teh Intel chipset. Games and Video apps just seem to run and play smoother with a Intel chip in the board.

---

To be honest, in most games, I doubt you can actually tell the difference. Generally the only affect is frame rate, and generally it only results in a 1-2 FPS difference, as evidenced here. When you're running an average framerate over 100, you generally aren't going to be dropping below 60FPS, and gameplay should be equally smooth.
As for "heavy processing", what precisely do you mean by that?

By heavy processing, I mean Video compression, Video editing, Zipping large files. I also use P4's when I'm optimizing a database etc.

In games chsh, I look at the FPS a lot and I do get more FPS with Intel than what i do with AMD in most games.