Hi, I don't know if this is helpful, but the DOSidle utility claims to enhance Media GX processors.

There isn't any. The MediaGX is a cut-down Cyrix6x86MX which is already a slow processor for most 3d games to begin with.

yep, but it should be enough for Doom at least.
Weird thing is that heretic (and even quake 1) Works much faster than Doom (the timedemo with doom is broken since it's slower when you enable timedemo)

Isn't MediaGX actually using a form of Cx5x86? I think I remember seeing Diablo 1 stutter on one.

What speed MediaGXm are you using? Benchmark results for all GXm speeds are reported in the Ultimate 686 Benchmark Comparison. For example, the FPU ability of a GXm-266 is about that of a Pentium 133, while the ALU ability is similar to that of a Pentium 166. On the other hand, if you look at Quake (software mode), the GXm-266 equates to a Pentium 100. For Quake II, it is similar to a Pentium 120. The DOOM results for the GXm-266 are similar to a Pentium 200 or a Pentium MMX 166.

I found the integrated graphics to have horrendous quality, so I went with a PCI Banshee card. The integrated audio also sounded terrible, so I went with an ISA sound card. There is an ultra DMA driver, which works in Win98SE, but it was a pain to get it working. Ultimately, I use a PCI Ultra100 card.

There are some performance tweaks for the GXm. I have outlined them here, Register settings for various CPUs
One such tweak is enabling FP_FAST, which helps the FPU results. For example, DOS Quake scores went from 33.1 fps to 36.2 fps after enabling FP_FAST on the GXm.

Basically, the GXm has speeds similar to the Cyrix 5x86, but are about 10% slower on account of not having any L2 cache. What the GXm has going for it is MMX. What I found surprising was that they didn't bother to fix branch prediction, not even with a GXm at 266 MHz. Yet branch prediction works even on the 6x86-P90.

I had a feeling Feipoa would provide the knowledge. 😀 Interesting to read about the integrated audio, video and DMA issues. And the still broken branch prediction. Strange indeed.

the processor is the :Cyrix MediaGX GXm-200BP 85C
I tried to use the set5x86.exe but it Always freezze at some point, no matter if I do what you did or if I set everything to off/disabled 🙁

Forget about set5x86 for now. You can use CTCHIP34 or 6x86ctl to turn on FP_FAST. That is the most important feature anyway.

Quake 1 scoring 33fps? here it's doing 21.4fps 😮
I tried both tools, but neither made a single 1fps change, maybe the gpu is too slow for noticing some change? (I saw it does 8MB/s on speedsys, that sounds like quite slow) (I belive something like a 16bits isa vídeo card)
EDIT: Actually, looking at speedsys, the cpu score didnt changed (94.90) but the gpu speed changed (17MB/s)

ibm5155 wrote:

Quake 1 scoring 33fps? here it's doing 21.4fps 😮
I tried both tools, but neither made a single 1fps change, maybe the gpu is too slow for noticing some change? (I saw it does 8MB/s on speedsys, that sounds like quite slow) (I belive something like a 16bits isa vídeo card)
EDIT: Actually, looking at speedsys, the cpu score didnt changed (94.90) but the gpu speed changed (17MB/s)

I am running Quake at 320x200 in software mode, so I'd be very surprised if you didn't notice any change. I suspect you weren't successful at enabling FP_FAST. Check to ensure it is enabled after enabling it. CTCHIP34 should verify this. You can also use the CTCHIP34 GUI to enable the feature and to determine which features are currently set. You can also use 6x86ctl.exe -q to see which features are enabled, but you must look at the register table. It is easier to use CTCHIP to determine what values are set.

It is possible that some of my settings for my system are not stable with your system. To manually enable just FP_FAST, go into CHCHIP's GUI,

CTCHIP34 cx586
Hit Page Down until you arrive at E8h
P
5=1
Hold down PG DN to exit.
Retest Quake.

You can also use 6x86ctl.exe to just enable FP_FAST. Use this command,
6x86ctl.exe -n4 -b20

feipoa wrote:

It is possible that some of my settings for my system are not stable with your system. To manually enable just FP_FAST, go into […]
Show full quote

It is possible that some of my settings for my system are not stable with your system. To manually enable just FP_FAST, go into CHCHIP's GUI,

CTCHIP34 cx586
Hit Page Down until you arrive at E8h
P
5=1
Hold down PG DN to exit.
Retest Quake.

You can also use 6x86ctl.exe to just enable FP_FAST. Use this command,
6x86ctl.exe -n4 -b20

Well, I tried, it's still near to 20 - 24fps home
Doom score with time demo was also 24fps
disabling the áudio from bios resulted in 30fps for doom and 21.1fps for quake1

setting the vídeo ram from 2.5MB to 1.5MB resuled in no performance gain.
disabling vídeo shadow also resulted in no fps gain.
cas ram 2 -> 3:lost almost one fps with that on doom

Are you telling me that with enabling FP_FAST (6x86ctl.exe -n4 -b20), you got 0.0 fps increase in DOS Quake? What is the timedemo1 result before enabling FP_FAST? What is the timedemo1 result after enabling FP_FAST?

Quake 1 performance: (with áudio disabled by bios)
FP_FAST disabled: 21.5FPS
After FP_FAST enabled: 21.4FPS
I'm using this comand for starting quake (quake -nosound -noipx -nojoy -nocdaudio +stopdemo +timedemo demo1)

Lookint at CTCHIP34 cx586, it says that FP_Fast is enabled.

More information about the processor:
-CPU-ID: 000540

VERY INTRIGUING! I am stumped. I have noticed that if the L1 cache is disabled, or if AISP is enabled (bit 3 of PCR), that most other register enhancements have little performance benefit. While it is highly unlikely, can you ensure that AISP is disabled? You can view this with CTCHIP.

It is also possible that your particular CPU has FP_FAST locked in the off state. I know a few of the Cyrix enabling programs claim FP_FAST is locked. CTCHIP may be falsely reporting it as enabled.

I wonder if there is something up with your onboard graphics. Do you have a PCI slot?

Not actually, well, maybe with some pcmcia to pci converter (netbook) 🤣

Here's the default CTCHIP log:

1L1-Cache CD: Enabled
2L1 Cache ND, may be locked by C2h:2: Write Throught
30/1 Alignment Check: Disabled
40/1 Write Protected: Enabled
5Index=C1h Configuration Control Register 1
6C1h 7654 0000 reserved
70/1 MMAC SMI-Service to main memory access: disabled
80/1 SMAC SMM Access with /SMADS output aktive: disabled
90/1 SMI SMM, otput pin /SMADS, input /SMI: enabled
10reserved
11
12
13Configuration Control Register 2
140/1 (SUSP) Suspended Pins (/Susp und /Suspa): enabled
150/1 (BWRT) Burst Write Cycles: disabled
16reserver
170/1 (WT1) Meomry access to 640...1M Always write thru: disabled
180/1 (HALT) Suspend on Hald: Enabled
190/1 (LOCK) Prohibit changing of the NW-Bit in CR0: Disabled
200/1 (WBAK) Write Back Cache Interface Pins: Disabled
21reserved
22(MAPPEN) Sleect Active Config Register Set for D0h .. FDh: Default
230/1  (SMIAct) Intel Compatble SMM (486SL): disabled
240/1  (LINBRST) Linear address sequence for Burst Cycls: enabled
250/1  (NM1) during SMN: disabled
26(SMI-LOCK): disabled
27reserved
280/1 (FP_FAST) Fast FPU exception reporting: disabled
290/1 (DTE_EN) Directory table entry cache: enabled
300/1 (MEM_BYP) memory bypassing: enabled
31I/O recovery time: 0 bus clock cycles
32I/O self modifying code checking: enabled
33reserved
34core clk: normal
35Clock mode Core/Bus
36Load/Store ordering: Strong
370/1  BTB Test Register: disabled
38reserver
390/1  Reordening of locked misaligned loads: disabled
400/1  All instructions stalled to serialize pipeline: disabled
410/1  Loopmode: disabled
420/1  Branch Target Buffer: disabled
430/1  Return Stack: disabled

EDIT: Tested PCBench, it gaves 28.3fps

https://www.youtube.com/watch?v=TCehzK9D8CI
Ok, here's the vídeo showing that actually, there are game that Works/sound fine and others that are meh, for some reason every game that I tried to benchmark got fps lower than the real game had. (Quake 2 and doom)

EDIT: Using the MBF port for running doom, made the fps jump from 21 to almost 50FPS !!!
I even found a dos port of quake 2, and it's doing 11,3FPS on ms-dos x 9,9 over Windows;
maybe with some special vanilla port of quake 1 I could reach 33fps like you did?

feipoa wrote on 2016-06-02, 09:27:

What speed MediaGXm are you using? Benchmark results for all GXm speeds are reported in the Ultimate 686 Benchmark Comparison. […]
Show full quote

What speed MediaGXm are you using? Benchmark results for all GXm speeds are reported in the Ultimate 686 Benchmark Comparison. For example, the FPU ability of a GXm-266 is about that of a Pentium 133, while the ALU ability is similar to that of a Pentium 166. On the other hand, if you look at Quake (software mode), the GXm-266 equates to a Pentium 100. For Quake II, it is similar to a Pentium 120. The DOOM results for the GXm-266 are similar to a Pentium 200 or a Pentium MMX 166.

I found the integrated graphics to have horrendous quality, so I went with a PCI Banshee card. The integrated audio also sounded terrible, so I went with an ISA sound card. There is an ultra DMA driver, which works in Win98SE, but it was a pain to get it working. Ultimately, I use a PCI Ultra100 card.

There are some performance tweaks for the GXm. I have outlined them here, Register settings for various CPUs
One such tweak is enabling FP_FAST, which helps the FPU results. For example, DOS Quake scores went from 33.1 fps to 36.2 fps after enabling FP_FAST on the GXm.

Basically, the GXm has speeds similar to the Cyrix 5x86, but are about 10% slower on account of not having any L2 cache. What the GXm has going for it is MMX. What I found surprising was that they didn't bother to fix branch prediction, not even with a GXm at 266 MHz. Yet branch prediction works even on the 6x86-P90.

Hi,

Regarding the UDMA driver, did you ever get it to work? I am testing an IGEL-4 thinclient with board MB-5BLGP its the same as a neoware eon. I am using a 44 pin IDE hdd. When i fisrt updated the driver it gave me conflict in device manager and hdd stop showing on device manager under the disks section. Then since this board only has 1 ide connector i set the controller driver settings to only use the primary IDE, rebooted and conflict went away, this disk showed up again and gave me the option to enable DMA. After enabling and rebooting the system hangs on the windows 98 boot screen, rebooting again and pressing escape there are no messages just hangs with blinking cursor. thanks.

his.lordship wrote on 2023-07-19, 13:59:

feipoa wrote on 2016-06-02, 09:27:

What speed MediaGXm are you using? Benchmark results for all GXm speeds are reported in the Ultimate 686 Benchmark Comparison. […]
Show full quote

What speed MediaGXm are you using? Benchmark results for all GXm speeds are reported in the Ultimate 686 Benchmark Comparison. For example, the FPU ability of a GXm-266 is about that of a Pentium 133, while the ALU ability is similar to that of a Pentium 166. On the other hand, if you look at Quake (software mode), the GXm-266 equates to a Pentium 100. For Quake II, it is similar to a Pentium 120. The DOOM results for the GXm-266 are similar to a Pentium 200 or a Pentium MMX 166.

I found the integrated graphics to have horrendous quality, so I went with a PCI Banshee card. The integrated audio also sounded terrible, so I went with an ISA sound card. There is an ultra DMA driver, which works in Win98SE, but it was a pain to get it working. Ultimately, I use a PCI Ultra100 card.

There are some performance tweaks for the GXm. I have outlined them here, Register settings for various CPUs
One such tweak is enabling FP_FAST, which helps the FPU results. For example, DOS Quake scores went from 33.1 fps to 36.2 fps after enabling FP_FAST on the GXm.

Basically, the GXm has speeds similar to the Cyrix 5x86, but are about 10% slower on account of not having any L2 cache. What the GXm has going for it is MMX. What I found surprising was that they didn't bother to fix branch prediction, not even with a GXm at 266 MHz. Yet branch prediction works even on the 6x86-P90.

Hi,

Regarding the UDMA driver, did you ever get it to work? I am testing an IGEL-4 thinclient with board MB-5BLGP its the same as a neoware eon. I am using a 44 pin IDE hdd. When i fisrt updated the driver it gave me conflict in device manager and hdd stop showing on device manager under the disks section. Then since this board only has 1 ide connector i set the controller driver settings to only use the primary IDE, rebooted and conflict went away, this disk showed up again and gave me the option to enable DMA. After enabling and rebooting the system hangs on the windows 98 boot screen, rebooting again and pressing escape there are no messages just hangs with blinking cursor. thanks.

I've had the same issue with a ECS P5GX-M. I'm guessing the reference drivers are hit and miss for DMA compatability and the OEMs modified them to work for their individual use case. I'd see if you could hunt down the driver package that shipped with your thin client.