NOTE:The following program and results have been taken offline!!
Introduction
The little ctcm program from the German ct-Magazine is an extreme helpful tool to find out about your motherboard performance. The most important thing to do for a motherboard is to transfer data as fast as possible from one device to the other. In particular this is the transfer process between CPU, L2 Cache, Main Memory and PCI Devices. Your CPU can be as fast as it wants, as long as your M/B isn’t able to get the data quickly from the memory or the graphics card to the CPU and vice versa, your computer will be as slow as a slower CPU in a fast M/B.
Exactly this is measured by ctcm. It checks the speed of your L1 Cache (which only is depending on your CPU speed really), of your L2 Cache – and here it checks all the different cache conditions – and of your main memory. You are also able to check the speed of the data transfer to and from your graphics card to learn about the speed of your PCI bus, but here it is also very much depending on your graphics card itself.
When you see the results, you have to take in consideration that one of the crucial things for your cache and main memory speed is the bus tact (or bus speed). As you know from my Pentium Reference Table, the bus speed is not the same in all Pentium computers, it depends on the CPU speed. A Pentium 120 has a bus speed of 60 MHz and has therefore a slower memory and L2 cache speed than a Pentium 100. A Pentium 133 has just as a Pentium 200 the same bus speed of 66 MHz – therefore the cache and memory speed is the same, except of the L1 cache.
How to use it
You have to use this program in DOS !!! No DOS-Box, no virtual mode device manager ! You’ll need an Extended Memory Driver though, HIMEM.SYS.
If you still should use Windows 3.1 (or WfW 3.11), you normally just have to leave Windows – as long as you didn’t load EMM386, QEMM or any other virtual mode device manager in config.sys. If you should have, ctcm will tell you. The probably easiest way to run ctcm is to create a boot disk with nothing in the config.sys. You won’t need a autoexec.bat, but if you’re smart, you put the line “C:ctcmctcm.exe” as only line in the autoexec.bat, where “C:ctcm” has to be replaced by your ctcm directory – wherever you store it on your hdd. In this case the computer will boot and straight away run ctcm.
In case you are a Windows 95 user (like me), you either anyway have a dual boot option to boot into old DOS, or you haven’t (tsts !!!). If you should have the boot menu, you just choose Start in MSDOS Mode. If not, you either press F5 continuously after the BIOS POST, cause that’ll get you into the boot menu, or you can get into MSDOS mode from Windows 95, via choosing Start Computer in MSDOS Mode from the shutdown menu or you’ll also have to create a boot disk as described above.
The real cracks have got a boot menu for their old DOS (as myself of course 😉 ) and certainly have an option to boot with only the XMS driver, as you need it for lots of the modern games nowadays. In this case it’s easiest of course, you boot this way and run ctcm, basta.
The results should speak for themselves, after all I eventually translated that nice little program ;-).
There are quite a few command line options for ctcm, but these I will explain at a later stage – or translate the readme file. You wont need any command line options for the results I would like you to send back.
How to get it
Currently not available
Well, that’s pretty easy – just click on this link: DOWNLOAD THE INTERNATIONAL VERSION OF CTCM !! NOW !!!
You only need to extract it and there you go !
How to send the results to me
Currently not available
These things I’d like to ask you for:
- Please tune up your BIOS Memory Timing before running the test. We want to see results of the mainboards as they perform best. An excellent board with lame BIOS settings performs crappie. Anyway this is a performance freaks site, so it’s a question of honour to adjust your boards to the limit, ain’t it?
- Please send in more results of boards with the Intel 430VX chipset. Boards with that chipset also seem to be able to run at a bus speed of 75 MHz, so try yours out! However under that circumstance the PCI bus runs at 37.5 MHz and is therefore above the PCI specifications. Anyway, there is a huge performance gain with a bus speed of 75 MHz, as long as it runs, and there’s a very low probability that you could cause any damage.
- Please send in results of boards with the VIA VP-1 or VP-2 chipset. The VP-1 is designed to support 75 MHz bus speed and it’s able to still run the PCI bus at 33 MHz – asynchronously though. I’d die for results with these chipsets. The only boards which use it so far seem to be from FIC.
- Please send in results of Pentium Pro boards!!!! I soon will put ct-magazine results of PPro boards on a new PPro boards page and I hope that I can back these up with a decent PPro board survey.
The Results
I created a PERL script, which generates a customized results table for you, to offer you the most convenient way to access the benchmark results. This kind of presentation is completely and unique on the whole World Wide Web. Drop me a line about what you think of it.
Well guys out there, the results are coming and it is just as I expected: as soon as you fine tune the memory settings in the BIOS, you’ll get good or even excellent results of motherboards, that didn’t perform too well in the ct-magazine test. Best example is the Supermicro. Another thing you definitely have to consider regarding the new 430HX boards: if you have a look at my chipset page you’ll see that the 430HX chipset is able to read EDO at a 5-2-2-2 and write at a 4-2-2-2 burst. This only works however with 50ns or better EDOs. The so called ‘Turbo Read Leadoff’ option in the BIOS setup is nothing else as doing 5-2-2-2 read bursts instead of 6-2-2-2. You will need these fast EDOs however (which I’ve already got ;-)), otherwise you’ll occur crashes. I only can tell you where I got my 45ns EDO, from ComputerNerd, USA. Don’t get any money from this place btw, but they are very helpful there, even to Europeans 😉 and therefore worth being recommended.
Discussion
We,ve got the results – what we gonna do with’em ?
First of all, don’t forget that all speeds except the L1 one depend significantly on the bus clock. A system with a bus clock of 60 MHz always has a slower memory speed than a system with 66 MHz bus clock. Therefore a P180 has a slower memory speed than even a P100. It’ll be very interesting to see, how systems with 75 MHz bus clock will perform, if they ever should turn up.
- L1 Cache Speed
As already said – this speed only tells us how fast the CPU speaks to its own chip cache, so it only depends on the clock speed of the CPU. Nice number, but it doesn’t tell us a lot at all. - L2 Cache Speed
This one is important. It tells us how fast the CPU is able to transfer data between itself and the L2 Cache. Highly dependent on Cache Type used – PB SRAM is favorite here. Also shows quality of M/B design, the faster, the better the M/B design – but don’t get fooled by small differences here ! More or less all boards are able to reach the physical limitation of 66.6 MB/s at 66 MHz clock speed – due to the 3-1-1-1 cache read cycle of all the chipsets using PB Cache. If you’ve got more it’s either due to measuring tolerance of ctcm or you might have a slightly higher clock speed than 66 MHz. Only 75 MHz bus speed boards can top that value. - Memory Speed
The most interesting one ! This value depends on three things:- the memory type used:
the faster the memory, the higher the value. EDO performs quite a bit better than FPM, and I’m waiting impatiently on results for BEDO and SDRAM, which should be even faster than EDO. - the Chipset:
One of the main differences between the 430FX and 430HX PCIset is the ability of the HX to access EDO RAM faster than the FX, the EDO timing is much better (max 5-2-2-2 for EDO in HX, max 7-2-2-2 for EDO in FX). There HAS TO BE a significant difference between the speed for FX and HX. It also would be very interesting to know how the new other chipsets (VIA Apollo Master, VLSI Lynx,….) perform here. - the RAM timing:
The less cycles it takes to write to DRAM, the faster the memory performance. You are normally able to adjust this in your BIOS setup – as long as you don’t run auto config. The values are RAM read, RAM write, Leadoff cycle, RAS to CAS delay. Keep them small to gain maximum performance but be aware of system crashes if your RAM isn’t up to it – this is the reason why I’m using 45 ns EDO.
Having said that, it’s up to the M/B manufacturer to make the most of it – so not every HX board is that much faster here than a FX board ! You should expect up to almost 40 MB/s for a FX at 66 MHz (my own I got tuned up to 39.3 !) and up to almost 48 MB/s for a good HX at 66 MHz. Have a look at the mainboard comparison table to see which boards seem to reach that. The reason why the HX chipset has to be faster is the different memory timing. The fastest read timing (with EDO) of the Triton FX chipset is 7-2-2-2, the fastest of the Triton II HX chipset is 5-2-2-2. Much faster even should be BEDO and SDRAM. Their memory timing in the Triton VX or the VIA Apollo VP is 5-1-1-1, so we should expect a speed of up to 57 MB/s !
- the memory type used:
This one tests the speed in the good old 640 kB block of main memory. Because this is very small, it’s mostly dependent on your L2 cache size and speed. Systems with 512 kB L2 cache are significantly faster than systems with only 256 kB, cause 640 kB fit almost completely in the 512 kB cache.
It’s tested using 4 MB memory to run the benchmark in, so it’s very dependent on your main memory speed. The cache plays only a small role here, for 4 MB are significantly bigger than 256 or 512 kB.
