==============================================================
Guild: wafer.space Community
Channel: Information / tinytapeout / This was a challenging one! After
After: 10/31/2025 23:59
Before: 12/01/2025 00:00
==============================================================

[11/22/2025 16:17] essen__



[11/22/2025 16:17] essen__
Nice, how did you guys go about changing the default cell library ?


[11/22/2025 18:07] rzioma
@Essen theoretically it should be just `"STD_CELL_LIBRARY": "gf180mcu_fd_sc_mcu9t5v0",` in `config.json`
BUT there is a bug in the PDK that TT is using, so @htamas had to make a patch: https://github.com/TinyTapeout/tt-gds-action/compare/main...rejunity:tt-gds-action-gf180mcu_fd_sc_mcu9t5v0:main


[11/22/2025 18:14] rzioma
So basically you have to
1) clone https://github.com/TinyTapeout/tt-gds-action apply the patch
1.5) add a tag on your patched gds-action repo, I used `ttgf0p2_9t`
2) route your gds.yaml to use patched gds-action repo. something like that: https://github.com/rejunity/tt08-vga-drop/commit/2ac440a4bd84136deccb49f6ed56c4ded1a0a8b7
3) add "STD_CELL_LIBRARY": "gf180mcu_fd_sc_mcu9t5v0", to your `config.json`
4) change Makefile so that Gate Level tests use correct cell library too: https://github.com/rejunity/tt08-vga-drop/commit/22135b09cfd3b970ce26098b89df31547c557192

and you should be done


[11/22/2025 18:16] rzioma
You might get around 10..15% speedup from that


[11/22/2025 18:16] essen__
Nice, sounds like it was a fun ride. 👍


[11/22/2025 18:16] rzioma
(but loose in area)


[11/22/2025 18:17] rzioma
for some definition of fun...


[11/22/2025 18:17] essen__
I think we all here share a common twisted definition.

{Reactions}
partyblob

[11/22/2025 18:18] essen__
Haha, maybe I should. 
For now I don't have any timing issues, at least, assuming the output io doesn't go faster than 50MHz ? 🤷


[11/22/2025 18:19] rzioma
I don't think this affects IO cells, only logic/dff


[11/22/2025 18:19] essen__
Yes, just to say that my timing is fine, and I am bottlenecked by the IO, so I have no motivation to up the frequency.


[11/22/2025 18:20] essen__
We need faster, open source IO so bad 🫠


[11/22/2025 18:21] essen__
Anyways, off to lunch 👋 see you around

{Reactions}
👍

[11/22/2025 18:21] rzioma
Yeah, for MAC-array/SystolicArray you bound to be IO bound, I suppose


[11/22/2025 18:21] rzioma
I added some "noob" input caches to my implementation, some time ago


[11/22/2025 18:22] rzioma
https://github.com/rejunity/tiny-asic-1_58bit-matrix-mul

{Reactions}
👀

[11/22/2025 18:23] rzioma


{Attachments}
2025-11_media/image-58839.png


[11/22/2025 18:25] rzioma
And septenary version: https://github.com/rejunity/tt09-septenary-matrix-mul
Unfortunately never got fab'd due to eFabless going bonk


[11/22/2025 18:56] essen__
Any plans on submitting it to a future shuttle ?


[11/22/2025 19:33] rzioma
Yes, for the next sky130 shuttle I would like to implementa input data cache with @tnt SRAM blocks.
I am also very tempted to see how large of an array could fit onto the whole wafer.space chip... maybe even buy a wafer for that 😉 but I don't think I will have time for the first shuttle.

{Reactions}
🤣

[11/22/2025 19:37] essen__
Sounds like more fun incoming 🙂 
Haha, an entire wafer, I would like to see you guys do that.


[11/22/2025 19:41] rzioma
ah, sorry, I didn't mean a wafer scale 😉 though that would be fun!!! Instead wafer.space has an option to get uncut wafer for $2000 with 40 chips of your design (but someone needs to cut and package them 😉)
but still not a bad deal because that gives approximately 128 TinyTapeout tiles worth of an area.

{Reactions}
👀

[11/22/2025 19:57] rzioma
Back of envelope calculation gives me around 10..20 GOPs (something like 16x20 mac array) for a whole chip on 180nm still far from 1TOps 🙂


[11/22/2025 19:59] essen__
@EmbeddedKen this is for you : 
> get uncut wafer for $2000 with 40 chips of your design (but someone needs to cut and package them 😉)


[11/22/2025 20:00] essen__
How did I miss that 👀 
Damn, looks like I also need a die bonding machine now.


[11/22/2025 20:01] essen__
How are you planning on doing the cutting and packaging ? Does one of you own the die bonding machine or know someone that does ?


[11/22/2025 20:02] 246tnt
Well the 2000$ option is only as an addition to the 7000$ slot.


[11/22/2025 20:03] 246tnt
So it's just to get a neat souvenir ... or if you want to do wafer level probing I guess.


[11/22/2025 20:09] essen__
So sad, you got my hopes up there for a second. 🤣


[11/22/2025 20:12] 246tnt
There is a fixed number of slots in the reticle ...  so the price is pretty much production_cost / num_slots ...  2k$ is the cost to run one more wafer through the flow.


[11/22/2025 20:13] 246tnt
The production costs are (simplified)  maskset_cost  + n_wafer * wafer_processing_cost    and the mask costs is the big chunks and the part which has a limited number of slots.

{Reactions}
👍

[11/22/2025 21:20] rzioma
Sad 😉

{Reactions}
🫂

[11/23/2025 18:59] essen__
Still a great deal though


==============================================================
Exported 34 message(s)
==============================================================