can everybody who has a K6-2 266Mhz - K6III+ 875Mhz please post screenshots of your results of cpumark99 and mem speed results in one of these: sisoft sandra,aida32,pcwizzard2004 or crystalmark99.also post what board you are using.

im using these to compare my system to other K6 socket 7 systems out there.


Thanks,Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

CPU: K6-III+ 450@560MHz
BUS: 112MHz
RAM: 1 x 256MB PC133@112MHz
CHIPSET: Via MVP3
MOTHERBOARD: SOLTEK SL-56H5
OS: Windows 98SE

SANDRA VERSION: 2003 Standard v1.91.31





Will try and send more ASAP Charlie - dam spyware is playing havoc with my machine at the moment..

-------------------------
. AMD CPU Data: http://www.tomshardware.com/20.../amd...ult/page23.html & http://www.amdboard.com/amdid.html
. Belarc Advisor: http://www.belarc.com/free_download.html
. GPU Comparison for Laptops: <a href="http://www.notebookche

wow,those mem speeds smoke mine,im curious as how the K6-2 500 and K6-2+ 550 smoke your K6-III+,im thinking that 1MbL2 but thats not full speed so whats up with that .will post my results using that software.
QUOTE Will try and send more ASAP Charlie - dam spyware is playing havoc with my machine at the moment..
Sygate Personal firewall seems to be keeping all of that stuf off my computer for now...


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

Hi Charlie,

Yes, I was also scratching my head wondering just why a CPU with less Cache would show an increase in cacheable data transfer. I also tested this CPU with exactly the same FSB and memory on that other FIC board I have - Via MVP3 but with 1MB L3 cache.

The memory bandwidth shot up to 196/192 but I didn't get a chance to test the cache-memory performance. Strange indeed as both boards had exactly the same memory timings and chipset drivers. Perhaps FIC just had a better design, as there is definitely a difference in overall performance with that board.

I know that these are synthetic tests, but your setup really seems to be holding you back. Maybe it could be down to the way the different buses are multiplexed between the onboard stuff and main memory and like my Soltek board, how SIS implemented their design. I'm not sure.

Best thing you could do is keep a look out for that super-duper SS7 setup and you'll be fine matey

-------------------------
. AMD CPU Data: http://www.tomshardware.com/20.../amd...ult/page23.html & http://www.amdboard.com/amdid.html
. Belarc Advisor: http://www.belarc.com/free_download.html
. GPU Comparison for Laptops: <a href="http://www.notebookche

OK here are my results just to compare.





one thing is clear,my cache smokes yours .
HMMM it dominates in the lower size blocks but is smoked when it gets to the bigger blocks...


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

Here's my Sandra scores... K6-3+ 450 @ 5.0x 115 = 575 MHz on my old Asus P5A


For comparison, here's the same mobo / CPU combo @ 6.0x 100 = 600 MHz


-------------------------
Here is my new PCChips M848A' ">http://img.photobucket.com/alb...2_T-2-3-3-6-2_M848.jpg v2.1, with a 1700+ @ 2411 MHz, and my trusty old PCChips M810L' ">http://members.lycos.co.uk/mmm...00MHz_Sandra_tests.jpg v7.1A, with a Barton 2500+ @ over 3500+. Some of my fancy cars are on this pag

Charilie,

According to this information, your SIS 530 has very poor RAM

QUOTE

http://www.stud.fernuni-hagen....142/...che.html#sis530' ">http://www.stud.fernuni-hagen....owto/cache.html#sis530

SiS 530
The cacheable area depends on the amount of cache memory, cache mode and the TAG-RAM width. This table is for 8-Bit TAG RAM and Write-Back Mode

512 kB 2nd-LC 64 MB
1 MB 2nd-LC    128 MB
2 MB 2nd-LC    256 MB



* More SIS 530 Data: http://users.rcn.com/chare/chipsets.htm' ">http://users.rcn.com/chare/chipsets.htm

More SIS 530 debate: http://www.stud.fernuni-hagen....142/...TML/000710.html' ">http://www.stud.fernuni-hagen....orum2/HTML/000710.html

Of course it may vary from board to depending on TAG-RAM configurations, but that's low. Plus, you're using a K6-III+ which should overcome this hurdle, but those memory benches are abysmul - they should be up there with my own and Milt's in treble figures, not double. You're running slower than PC66! (in Sandra, confirm your memory bus speed because there's some questioble data given in that table I have linked to above *).

The initial surge in cache-memory performance is down to the L2 cache inside the CPU, but as you've seen the moment this data resides outside that range (even using the L2 cache), it plummits. At 1MB chunks that data is being processed at a very low speed and it's no wonder your system feels sluggish with heavy loads.

You're running CTU or a similar program to enable write allocation, when you boot right?

I know it's a big chore inside that case of yours, but I'd be interested to see the benches with just 64MB or RAM..

So to summarise, I think your low RAM speed could be due to one or all of the following:

1) Shared bus
2) Small cacheable memory range
3) K6-III write allocation for entire memory not being fully implemented.

p.s. I also found that when posting images, using the .png format gives a lower files size than .jpg depending on the colours used.

-------------------------
. AMD CPU Data: http://www.tomshardware.com/20.../amd...ult/page23.html & http://www.amdboard.com/amdid.html
. Belarc Advisor: http://www.belarc.com/free_download.html
. GPU Comparison for Laptops: <a href="http://www.notebookche

I bring you back to the CacheChk data you ran for me...

Your board...100MHz Memory Bus -

L1 cache is 32KB--2501.5 MB/s 0.4 ns/byte (1993%)
L2 cache is 256KB--1673.9 MB/s 0.6 ns/byte (1333%)
L3 cache is 512KB-- 446.7 MB/s 2.3 ns/byte (355%)

Main memory speed -- 125.5 MB/s 8.4 ns/byte (100%) [reading] 19.2 clks
Effective RAM access time (read ) is 67ns (a RAM bank is 8 bytes wide).
Effective RAM access time (write) is 114ns (a RAM bank is 8 bytes wide).


My board...66MHz Memory Bus -

L1 cache is 32KB--1659.8 MB/s 0.6 ns/byte (983%)
L2 cache is 128KB--1109.0 MB/s 0.9 ns/byte (657%)
L3 cache is 512KB-- 369.9 MB/s 2.8 ns/byte (219%)

Main memory speed -- 168.7 MB/s 6.2 ns/byte (100%) [reading] 9.5 clks
Effective RAM access time (read ) is 49ns (a RAM bank is 8 bytes wide).
Effective RAM access time (write) is 113ns (a RAM bank is 8 bytes wide).


Look at the Main Memory speed...I'm hitting 169MB/s on 66MHz, while you are hitting 126MB/s on 100MHz. Effective read time...49nS vs. 67nS...9.5 clock cycles vs. 19 clock cycles!!!

So Wildcard is right on...you're running slower than PC66 (which is effectively what I'm running).

I would bet it is the shared memory archetecture slowing you down. If you have a video card...any one...slap it in, then run the tests. Make the video card your main display, disable the shared on-board one (if you can), and then run the CacheChk test again...see if things improve.

Wildcard - His BIOS is fully Write Allocating...see the CTU photo. I don't think it is his cacheable range, since my VX only caches 64MB...

-------------------------
KachiWachi

Moderator - Wim's BIOS

CPU #1 - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), 128MB EDO. BIOS patched by Jan Steunebrink.
CPU #2 - Amptron PM-7900 (M520), i200 non-MMX, 128MB EDO
CPU #3 - HP8766C, PIII-667, 768MB SDRAM
CPU #4 - ASUS P3V4X, PIII-733, 256MB SDRAM
CPU #5 - ??? ;)

Also check out this post on the PCChips Lottery concerning Latency...

http://www.stud.fernuni-hagen....142-...ic=3094&start=0' ">http://www.stud.fernuni-hagen....m=3&topic=3094&start=0

Peter Missel - "The RAM latency and throughput of Intel VX is faster than the SiS 5598 simply because the latter has a shared-memory graphics unit that eats into RAM bandwidth obviously - and in text modes, bandwidth consumption from VGA is quite massive, about as high as in hi res graphics modes (because the character images are also in RAM)."

And some results from "bill28099" on a M598 motherboard. The clock is set at 6x66 so the data would relate directly with my DFI board. Note this board has the SiS530 video chip.

K6-II/+ @400MHz (6*66MHz)
L1 cache R/W 3048 MB/s
L2 Cache R=1527 W= 1018
L3 Cache R=291 W=129
Main Memory R=52 W=32
256K byte latencies 2/24/84/468
Null Size: 2 cycles 1 cycle (overhead 34 cycles)


My DFI (K6-II/+ @400MHz (6*66MHz) -

L1 cache (32KB) speed (MB/s): Read=3033.7, Write=3033.7
L2 cache (128KB) speed (MB/s): Read=1520.2, Write=1013.5
L3 cache (512KB) speed (MB/s): Read=336.5, Write=146.8
Main memory speed (MB/s): Read=144.2, Write=70.1
Latency 128K: CPU Clocks: 2/23/72/166 Bus Corrected: 2/23/12/27.5
Null size: 2 cycles 1 cycles (overhead 41 cycles)

-------------------------
KachiWachi

Moderator - Wim's BIOS

CPU #1 - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), 128MB EDO. BIOS patched by Jan Steunebrink.
CPU #2 - Amptron PM-7900 (M520), i200 non-MMX, 128MB EDO
CPU #3 - HP8766C, PIII-667, 768MB SDRAM
CPU #4 - ASUS P3V4X, PIII-733, 256MB SDRAM
CPU #5 - ??? ;)

if you mean can i disable it in the bios,no in device manager yes,and i have already done this as well(see other thread) but will do this again with the 128 then the 64.will post results...


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

OK i have the results...

with my pc133 128 stick of ram:




the 64Mb stick:




the sis530 was disabled and was using the 1Mb s3trio64 (was 2 but the add in modules burnt out for no reason )
also i seem to remember KachiWachi to post something about something on my ram,will these do?:

the 64Mb stick is the one with the 4 chips (its double sided so thats 8 chips total).
the 128 chip is single sided.


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

I can't read the markings on the chips Charlie, but thanks for the photos anyway. Can you try again?

The one on the left has Micron Chips (MT), but I can't read all the numbers (MT48LC16M).

I wanted to compare your numbers, but the old pics aren't coming through right now, so I'll stop back later on.

-------------------------
KachiWachi

Moderator - Wim's BIOS

CPU #1 - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), 128MB EDO. BIOS patched by Jan Steunebrink.
CPU #2 - Amptron PM-7900 (M520), i200 non-MMX, 128MB EDO
CPU #3 - HP8766C, PIII-667, 768MB SDRAM
CPU #4 - ASUS P3V4X, PIII-733, 256MB SDRAM
CPU #5 - ??? ;)

Hey Charlie, that memory bench looks a lot better, although you're still being held back compared to other chipsets. Amazing the performance plunge once you go beyond 128MB, eh? This board doesn't take to large amounts of RAM.

Also, take those MMX results with a pinch of salt - they're mostly CPU-based but do illustrate the advantage of a faster FSB which helps keep the SIMD registers tanked up.

Hmm. Looks like you're in a no-win situation here, what with the nature of your computing habits (rendering!. You're stuck with low RAM and higher memory bandwidth that still strangles the system, or large amounts of RAM and abysmal bandwidth. It's a shame because of the CPU you're using, which is spending most of it's time crying out for data.

Are you using that sound card as well or still relying on the onboard stuff? Iv'e seen what it's like inside your case and undertsand if you don't exactly relish the thought of trying to reach the jumpers or make some extra space

-------------------------
. AMD CPU Data: http://www.tomshardware.com/20.../amd...ult/page23.html & http://www.amdboard.com/amdid.html
. Belarc Advisor: http://www.belarc.com/free_download.html
. GPU Comparison for Laptops: <a href="http://www.notebookche

@wildcard,I'm using the ALS400 you sent me and I'm loving it ,it doesn't make the sound skip when playing games,getting to the jumper is no problem with me and as far as space goes Ive rolled up the cables to be round and tied everything down to make as much room possible but still is like a coffin in there .
@KachiWachi,withoud thinking i deleted out the pics in my account,and because of file size limits i cant get the res any higher than that,will just try out scanning one chip instead of the whole thing.will get those pictures back up because i just re-uploaded them .


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

Just read me the chip numbers...I don't need a photo...

You should see changes at the breakpoints (data is from your board)...

L1 - 32kB
L2 - 256kB
L3 - 256kB (same place...so there is overlap)
RAM - 64MB (end of cacheability range)

CacheChk will show that best, as well as CacheMem and CacheGraf.

L1 - 0.4 us/kB
L2 - 0.7 us/kB
L3 - 0.7 us/kB (overlap again)
512K - 2.5 us/kB (some residual cache effects)
RAM - 8.8 us/kB

-------------------------
KachiWachi

Moderator - Wim's BIOS

CPU #1 - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), 128MB EDO. BIOS patched by Jan Steunebrink.
CPU #2 - Amptron PM-7900 (M520), i200 non-MMX, 128MB EDO
CPU #3 - HP8766C, PIII-667, 768MB SDRAM
CPU #4 - ASUS P3V4X, PIII-733, 256MB SDRAM
CPU #5 - ??? ;)

ok scanning didnt work,the numbers just wont show,so i wrote them down:
the 64Mb stick:

-8t
EliteMT 9836s

M12L64164A D6Y051D6

THE 128 STICK:

0136 L-1

MT 48LC16M8A2

TG -75E




i typed them the way they looked on the chips if it matters.

@KachiWachi,im curious as to what you are looking for that gives you the info...


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

Thanks for the numbers.

I wanted to see some info on the chips so that we might be able to better set your BIOS up for them.

A quick look shows both parts to be CL2@100MHz, so that is good.

Now I have to go look at what settings you have available...

-------------------------
KachiWachi

Moderator - Wim's BIOS

CPU #1 - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), 128MB EDO. BIOS patched by Jan Steunebrink.
CPU #2 - Amptron PM-7900 (M520), i200 non-MMX, 128MB EDO
CPU #3 - HP8766C, PIII-667, 768MB SDRAM
CPU #4 - ASUS P3V4X, PIII-733, 256MB SDRAM
CPU #5 - ??? ;)

so that means i should be able to pull 800Mb/s+ off of them then?
its storming here so I'm getting off for now,see you guys later


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500

OK...need some info...

---------------------------------------------------------
Refresh Command Delay-9,8,7,6
RAS Precharge Time-5,4,3,2
RAS to CAS Delay-5,4,3,2
Refresh Queue Depth-0,4,8,12
CAS Latency-2,3 T
Inital Latency Timer-off,32,24,16 Clocks
---------------------------------------------------------

I need units for these...except for the last two.

Also...need your settings for Trend ChipAway Virus, Share Memory Size, SMART, Boot OS/2, System/Video Cacheable, and Graphic Win Size.

Also, if you can post it...see what SAMDRA or Everest posts for your SPD data.

-------------------------
KachiWachi

Moderator - Wim's BIOS

CPU #1 - DFI 586IPVG, K6-2/+ 450 (Cyrix MII 433), 128MB EDO. BIOS patched by Jan Steunebrink.
CPU #2 - Amptron PM-7900 (M520), i200 non-MMX, 128MB EDO
CPU #3 - HP8766C, PIII-667, 768MB SDRAM
CPU #4 - ASUS P3V4X, PIII-733, 256MB SDRAM
CPU #5 - ??? ;)

here the are things in the *thing* are set:

trend chip away virus *enabled* disabled
share mem size 2M 4M *8m*
smart *enabled*
boot Os/2 *disabled* enabled
video 32k shadow *enabled* disabled cached
c000 16K ^SAME^
c400 16K ^SAME^
refresh command delay 9 8 7 *6*
ras precharge time 5 4 3 *2*
ras to cas delay ^same^
refresh queue depth *0* 4 8 12
graphic win size 4mb 8mb 16mb 32mb *64mb* 128mb 256mb
cas latency *2t* 3t
initial latency timer *off* 32 24 16 <-clocks

wasn't sure what you meant but i think i got it down.
let me know if there is any other info you need, i hope this is what you needed.


Charlie.

-------------------------
Desktop:
Phenom II x6 1055T @ 4GHz | 4x2GB Patriot DDR3 1600 @ 2000 | 3x AMD Radeon HD6970 Crossfire

Laptop:
Core i7 2960xm @ 4.2Ghz | 4x4GB Kingston DDR3 1866 | 2x GTX 580m SLI OC 725/1450/1500