UCA - Now testing early DRAMs like 4164/41256! [2022-02-07]

[xsecret]

Updates:
- 2020-04-03 : Added support for Intel 486s
- 2020-04-08 : Added support for AMD 486s & 5x86s
- 2020-04-21 : Added support for Cyrix, IBM & ST 486/5x86s
- 2020-04-28 : Added support for Texas Instruments 486s
- 2020-05-01 : Added support for UMC 486s
- 2020-07-02 : Added support for Intel 386DXs
- 2020-07-28 : Added support for AMD & IBM 386s
- 2020-07-31 : Added support for Intel RapidCAD
- 2020-08-21 : Added support for Cyrix 486DLC/DRx² & TI 486SXL(2) (PGA132)
- 2020-09-30 : Added support for Intel 286
- 2020-10-19 : JTAG Support Added to the UCA 486 Adapter
- 2020-12-11 : Added support for Intel 80188/80186
- 2020-12-28 : Added support for Intel 8087 FPUs
- 2021-01-19 : Added support for Intel 487 SX
- 2021-04-02 : Final Universal Chip Analyzer disclosed!
- 2021-06-18 : Experimental support for Pentium Overdrive
- 2022-01-25 : Added support for C&T Super386
- 2022-02-07 : UCA now able to test early DRAMs! (4164/41256)
-

-----------------

The whole story is probably too long for a thread here, so I wrote a post on my blog about the PGA Shield and the i486 support.

A log of progress has been made, and I'm now confident the base platform (FPGA PCB + the new PGA Shield) will soon be able to support everything from 4004 to 486/5x86!

Here is the story:

https://x86.fr/universal-chip-analyzer-supports-486/

All updates will be posted here. I'm already working on many adapters, so expect news soon!

Any question or suggestion, please ask!

[wren4777]

This is awesome!

[Vegeta]

Cool! For me please too

[CPUShack]

Its a pretty cool setup for sure

[debs3759]

Cool! I'm still hoping you create a motherboard that they will all run in Never did find motherboards for the 486 SXL

[xsecret]

Good news : AMD 486 & 5x86 are now supported!

If you need additionnal screen shot, just check my blog post here :

https://x86.fr/the-uca-486-adapter-now-supports-amd-486-5x86/

Here are the CPU tested :


Am486DX2-50 (1) - Am486DX2-80NV8T (2) : 8 KB L1 Write-Through Cache. Both CPUID 0x432 with cpuid instruction not supported. Virtually undetectable from Intel DX2 : Same microcode. Same power consumption. Exact same performances. Maybe distinguishable from Intel DX2 with JTAG. Work in progress on this point.

Am486DX4-100NV8T (3) : 8 KB L1 Write-Through Cache. CPUID also 0x432 without cpuid instruction. Real nightmare to distinguish from Intel. B-Step a bit lower power (-5%) Vs AY-Step. INT Perfomance is lower than Intel DX4 (-20%). FP Performance almost identical. Power consumption is also lower (-25%). One interesting thing : In 2x mode, the CPU is exactly as fast as an Intel DX2. However, in 3x mode, it seems significantly slower than a DX4, but only in INT. Performances looks like 2.5x in INT (according to cycle count) and "real" 3x in FP. Strange. That deserves some additional investigation.

Am486DX4-120SV8B (4) - Am486DX4-100SV8B (5) : Newer core with SMI and Write-Back L1 Cache (still 8KB). CPUID instruction supported. 2x/3x Mode and WT/WB change CPUID (0x434/0x474 in 2x mode, 0x484/0x494 in 3x mode). Exact same performance than NV8T. Bit higher power consumption (+10%).

Am5x86-P75 (Am486DX3-133V16BGC) (6). 16KB Write-Back L1 Cache ! CPUID instruction supported. 3x/4x Mode and WT/WB change CPUID (0x484/0x494 in 3x mode mode, 0x4E4/0x4F4 in 4x mode). INT Performance identical to 486DX4-100 (looks like INT units are locked in 3x mode). FPU performance are far better (real 4x). Doesn't support overclocking to 160 MHz.

Am5x86-P75 (Am486DX3-133W16BGC) (7). Same than previous one, but supports overclocking to 160 MHz with 3.6V !

[cvandijk]

Interesting. Are you planning on testing a A80486DX4-90NV8T?

And is it possible to run custom x86 code on it? Or write your own firmware slot to support x87 FPU's or other CPU's?

[xsecret]

The A80486DX4-90NV8T shouldn't be different than any other NV8T, but yeah ! I think I have an ES. Some Rare CPU / ES Analysis with the UCA is planned with a new dedicated thread, soon

Yes, you can run any x86 code in it. The toolchain is available and the code must be compiled with NASM, then it's possible to read/write memory from Windows to get results. That's quite easy.

It's also possible to write your own HDL code for any CPU. The pinout is open, I can release the .ucf and the Xilinx Toolchain (ISE, actually) is free. But that's much harder if you're not not fluent in Verilog or VHDL.

[cvandijk]

Don't have experience with Verilog or VHDL, I was hoping for something more simple, like this pin is for this, and that pin for that.

Nice to hear that we can run custom x86 code on it, that means we can write our own benchmark code or something completely else, like random code to detect unknown CPU bugs.

[xsecret]

Absolutely. I will soon ask some help about the benchmark code. It's not easy to define the correct mix of instructions to compare CPUs from 70s to 90s.

[Neology]

Awesome! I am definitely in if it is for sale

[xsecret]

Will be soon. But I need to finish some adapters first, then I'll be able to panelize different adapters on the same PCB to save on costs.

[Alitoutou]

cool! I want it
I'm more concerned about whether it can test AMD/cyrix/ST/UMC 486&5x86 finally.Wish you every success.

[cvandijk]

Are you going to create a Kickstarter for those?

[CPUShack]