Page 2 of 2

Re: World's smallest acceleator card atm

Posted: Sun Oct 06, 2019 1:56 pm
by utri007
Vampire core has some problems, it is not created to keep up compatibility, but add new features.

Other cores are better from my point of view.

Re: World's smallest acceleator card atm

Posted: Sun Oct 06, 2019 4:10 pm
by 8 Bit Dreams
Yes, indeed, Majsta started his project from 0, it is always hard to make first steps..this isn't something ready to go, rather something to start from.. Of course caches are not working, no mmu support and many more things to optimise, yet something we can use just right now...

Re: World's smallest acceleator card atm

Posted: Sun Oct 06, 2019 4:16 pm
by Cyprian
8 Bit Dreams wrote: Sun Oct 06, 2019 10:02 am well.. it was mentioned that this will be open-source, so if these lines are not connected - it will be possible to add them :!:
thats cool news

Re: World's smallest acceleator card atm

Posted: Wed Oct 09, 2019 12:42 am
by terriblefire
Id love to be able to put BGA on my accelerators...

But people wouldn’t be able to build them.

I’ve never done BGA.

Re: World's smallest acceleator card atm

Posted: Wed Oct 09, 2019 9:09 am
by adam
terriblefire wrote: Wed Oct 09, 2019 12:42 am Id love to be able to put BGA on my accelerators...

But people wouldn’t be able to build them.
Try us:)

Re: World's smallest acceleator card atm

Posted: Thu Oct 10, 2019 12:26 pm
by terriblefire
I wouldn’t be able to reliably solder a BGA.

Re: World's smallest acceleator card atm

Posted: Thu Oct 10, 2019 12:57 pm
by go0se
I spent a significant amount of time, effort and money a couple of years ago attempting to master the art of BGA work. Admittedly it was with board repairs, and dealing with significantly larger chips than those pictured on that accelerator.

The outlay for decent IR equipment that will give you accurate and repeatable temperature profiles is significant. The "Heatgun and Hotplate" approach just won't cut it when the narrow temperature ranges required can mean the difference between success and chip/board delamination.

The element of trust and guesswork that just can't be completely eliminated when working with BGA when you are unable "see" your results also didn't sit well with me.

I found the experience fairly soul destroying and ended up selling all my BGA equipment and vowed never to attempt it again, I have great respect for those who can do it but I'm not ashamed to admit that it's beyond my capabilities to achieve consistent satisfactory results.

SMT soldering on the other hand is a pleasure! :)