luckybob wrote:

I'm sitting outside in my workstation now that its FINALLY under 75F outside. (its 12:30 am) I have a complete setup ready for windows for slot A. I also have about 5 processors I can test. from what I see off hand... I have a 5, 6, 7, 950 and 1ghz slot A. I also have 5,6,667,850,1g pentium 3's...

speedsys scores would probably be "easiest" as I wont have to dick around with windows, however I dont think its a very good comparison. any suggestions for one or 2 benchmarks? 2 cpu marks and a video? to see if the cpu makes a difference in game fps?

I hate the heat my self, tired of 105+ that it has been for the past few days and it makes me tired 😵

Here are some scores that my rig is getting (+/- 100pts for random) variation

2001 default everything 3084
2000 default everything 5848
1999 3d default everything 6319
1999 cpu default everything 13046

Yeah; For classic Athlons vs "the others" like Duron, Thunderbird, Coppermine. and maybe early Willamettes Was there any big difference in applications that mattered to us at the time (games anyone? pretty sure this is a website dedicated to classic gaming; and not scientific applications)

Slot A Athlon and Coppermine are very evenly matched. Duron performs like those two. Tbird was usually about 10% faster.

There are some cases where Willamette would step ahead from its usual mediocre position. Video encoding gets along well with its architecture. Also, Quake3 likes it (link)

sprcorreia wrote:

Next in line for testing is:

Socket 370 P3 1GHz FSB133 vs Socket A Athlon 1GHz.

I'll get results for this whole 1GHz mess. 😁

awesome! what chipsets are you using?

because I'd like to see something comparing the irongate and the VIA KX-133. I know I have the latter, and it has overclocking options OUT THE WAZOO, not to mention support for inter weaved pc133 and agp 4x. it supposedly will let you set your own cache divider, but i havent played with it yet.

I've moved recently and the new apartment is like micro ATX size, so no space for all my rigs. I'll try to run those tests but it may take a while to do it.

swaaye wrote:

Also, Quake3 likes it

But who was still gaming at 640x480 on a P4 system? Step up to 1024x768 and any noticeable advantage the P4 may have had over the competition disappears. Almost everyone I knew at that time had a video card and monitor that could handle that resolution. We were already up to GeForce 2 and Radeon R100 at the time the review that you linked to was published.

I run most games at 640x480/800x600 on my P4 rigs 🤣

mr_bigmouth_502 wrote:

I run most games at 640x480/800x600 on my P4 rigs 🤣

Why? A GeForce2 GTS can give 50+ FPS in most of the games of the day in 1024 easily. Why settle for lower image quality than your system is capable of?

I mean, look at this

65 FPS in 1600x1200 in Q-III Arena on a P-III 933. Why would you even bother playing in 640 or 800?

sliderider wrote:

But who was still gaming at 640x480 on a P4 system?

When you want to compare how well CPUs run a 3D game, you reduce resolution so the video card isn't what's being benched.

The faster CPU may very well offer better performance at higher resolutions in the form of more stable frame rate. Old hardware reviews never looked at this though. Some modern reviews do this though, like at Tech Report and HardOCP.

swaaye wrote:

sliderider wrote:

But who was still gaming at 640x480 on a P4 system?

When you want to compare how well CPUs run a 3D game, you reduce resolution so the video card fillrate isn't the primary influence.

But you then make the testing unrealistic because given the choice of playing games in the lowest possible resolution with the fewest features turned on and playing in the highest possible resolution with more features turned on, most players will run the games so that it looks as good as it possibly can while still maintaining an acceptable frame rate. Playing a game in 640 with everything turned off so you can get 200 fps on a monitor that only has a 65hz refresh rate to start with is pointless. You're better off to sacrifice some of that excess fps to make the game look better.

sliderider wrote:

But you then make the testing unrealistic because given the choice of playing games in the lowest possible resolution with the fewest features turned on and playing in the highest possible resolution with more features turned on, most players will run the games so that it looks as good as it possibly can while still maintaining an acceptable frame rate.

There is the consideration of minimum frame rate, and of frame rate stability in general. Say when you have 10 Quake3 models on screen at once for an instant and the CPU load is much higher than the usual empty room. You would rather have the Athlon 1.2 than a P3-500, even if some 1024x768 charts show they are supposedly equal.

Some modern reviews are graphing frame rate stability in CPU and GPU benches. It's a better indicator of tangible performance than an average.

historically, AMD cpus have always been slower than intel cpus when it comes to PI calculation, if you don't believe me have a look at any forum that has a superpi 1m ranking section and have a look at the results..

As an owner of the very first 1GHz CPU (the "real" 1GHz with the K75 core, not the lame Thunderbird OEM version) let me put my 2 cents:

The 1GHz Athlon is a legend! Of course CPUs with on-DIE L2 cache perform faster due to full speed cache and much lower latencies. But they are also faster because the chipsets were way more efficient than the AMD and die VIA Slot-A chipsets. Many (VIA based) boards had trouble running the RAM in async mode of 133MHz, many didn't benefit from this due to bad BIOS implementation. But the CPU was stronger than the Katmai and still competed very well witch the early (sub 1GHz) Coppermine CPUs. Gaming with this CPU is Fun. It's just not all about superPI. Give it a Geforce DDR or a Fury MAXX (yeeha 😉) or maybe a Voodoo 5 -it rocks and brings back all the memories of the Y2K games.
Of course in 1:1 clock comparison the Coppermines were faster due to the more advanced technology. The 1GHz Athlon was just the very last blast from a already dying architecture.

Overall the Slot-A platform was very successful and AMD gained marketshares from Intel and funds to invest in the development of the next generation, the Thunderbird.

At the bottom line this is a very prestiguous and important CPU, shining bright for AMD. It's computer history.
Period.

😉

I'm still rocking the Barton core Athlon 2800, it's been through three lightning strikes/power surges that killed three motherboards and it still works. I was having a fun time with Unreal Tournament 2004/Chaos Mod the other day. Proxy mines! 😀 I really like that processor.

My voodoo 2 and 5 rigs are amd chips. 1ghz slot A and 2x 2800+'s respectfully. Amd has always been #2 in my mind, but they are always #1 in price/performance ratio. My current main system is a dual socket F. It would cost me DOUBLE to build an similar intel system, and I would only get 10% more performance.