Quake3 biased Against AthlonXP cpus ! Large speedups possible!

rms

Member
May 5, 2003
46
0
0
http://speedycpu.dyndns.org/opt/

I think this guy is worth talking to. He argues that Quake3 is very biased toward Intel chips and if it simply detected the athlon as SSE-enabled rather than 3dnow-enabled you would see huge speedups.

He apparently worked with AMD trying to incorporate this into the Q3 codebase, and Carmack refused for some reason. Anyway, this is worth a story !

I've talked to a couple of people who worked on adding 3dnow support to Quake, and if I understand them correctly, there is no 3DNOW/SSE support at all for the Athlon series of processors in Quake3. If correct this seems like a massive omission on ID's part that would bias Q3 benchmarks towards Intel chips bigtime.

rms
 

Megatomic

Lifer
Nov 9, 2000
20,128
6
81
Well, his optimization pack is worth trying. I just downloaded it and I'll try it out later.
 

rms

Member
May 5, 2003
46
0
0
I'm also wondering about the Athlon64 and Opteron in Quake3. If its true that there is no SSE support in the game for these cpus, this is a massive handicap that makes such benchmarks unfair in the extreme.

rms
 

Wingznut

Elite Member
Dec 28, 1999
16,968
2
0
Q3 is hardly SSE2 optimized. The reason that it performs so well on the P4 is due to memory bandwidth. The results that Hammer is having with Q3 proves that.

Yes, if programmers were to optimize their product for the Athlon, you would see some benefits. However, it's not going to happen very often that a developer wants to invest in a project that might benefit ~10% of the market.
 

rms

Member
May 5, 2003
46
0
0
I said SSE, not SSE2. This feature is there already in the Quake3 code, and the Intel chips are allowed to use it, while apparently no Athlon SSE-capable cpu is allowed to use it. Sounds unfair to me.

rms
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
I would be interested in seeing benchmark results for this. I have often wondered just how much of a "boost" SSE and the like give to applications and such. In my experience, MMX did next to nothing, SSE did a little bit more than nothing, and now with SSE2 people are saying that it actually has an effect. IMHO it's just intel putting their grasp on the hardware community by patenting some silly "optimizations" that aren't really necessary for good system performance.

Anyone got any numbers? My guess is less than a 2% gain from SSE over not using it in Quake III.
 

JeremiahTheGreat

Senior member
Oct 19, 2001
552
0
0
Originally posted by: Wingznut
Q3 is hardly SSE2 optimized. The reason that it performs so well on the P4 is due to memory bandwidth. The results that Hammer is having with Q3 proves that.

From a discussion @ aceshardware http://www.aceshardware.com/forum?read=105035344 , it would appear that it is because of lower latency caused by having large amounts of memory bandwidth (hardware prefetch). Athlon 64 single-channel seems to be kicking Opteron in Quake3 according to rumours.. with half the bandwidth but with reduced latency due to using non-registered, non-ecc ram, CL2 ram.
 

Accord99

Platinum Member
Jul 2, 2001
2,259
172
106
Out of curiousity, I tried the DLL on a P4. I get 35 more fps on the same demo and settings he used for his benchmarks. Most likely the gains come from using the latest Intel compiler, which generates the best performing code for modern x86 CPUs.
 

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
28,818
21,566
146
Originally posted by: Magdalene
interesting. anyone knows how to benchmark on quake? i've forgotten.

apparently mine still has the frame counter running at all times, but that doesn't help since it only goes up to 99 and is stuck there.


gottta love q3 though, no need to reinstall, my old install still worked. self contained.

sadly theres no reference point for most of us by feel atleast. it was silky smooth before all my upgrades, and silky smooth now
To benchmark it just hit the ~ key and type /timedemo 1 <enter> demo four <enter> I also suggest using the Extreme Quality settings which Thugsrook provides here
 

SickBeast

Lifer
Jul 21, 2000
14,377
19
81
Originally posted by: Wingznut
Originally posted by: rms
I said SSE, not SSE2. This feature is there already in the Quake3 code, and the Intel chips are allowed to use it, while apparently no Athlon SSE-capable cpu is allowed to use it. Sounds unfair to me.

rms
Well, if that's the case, then the P3 should show significant gains over the Athlon... But it doesn't. The Athlon 1.2ghz is faster than a P3 1.2ghz in Q3.

The P3 doesn't show significant gains over the Athlon because of its strong FPU. You seem very sceptical, but there may in fact be something to this argument that SSE would boost AMDs Quake III scores.

Intel=MMX, SSE, SSE2
AMD=brute force and a little bit of 3Dnow
 

Wingznut

Elite Member
Dec 28, 1999
16,968
2
0
Originally posted by: SickBeast
The P3 doesn't show significant gains over the Athlon because of its strong FPU. You seem very sceptical, but there may in fact be something to this argument that SSE would boost AMDs Quake III scores.

Intel=MMX, SSE, SSE2
AMD=brute force and a little bit of 3Dnow
I'm skeptical because SSE is not the reason that the P4 is so adept at Q3... Like I said above.

And even if this .dll file does improve the Athlon's P4 score, it is not a controlled comparison... We have no idea what this .dll does, especially in light of Accord99's increased scores using this .dll with his P4.


 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |