lekernel | hi | 12:13 |
---|---|---|
lekernel | what do you think about a name change to "milectra"? | 12:13 |
lekernel | sounds more generic than "milkymist" - milkdrop stuff is hardly what works best here. and less strange, too. | 12:14 |
juliusb | lekernel: for the entire project? | 15:14 |
lekernel | yes | 15:16 |
juliusb | sounds like the name of a company which installs electric fences around military installations :P | 15:17 |
Fallenou | I kind of like milkymist :p | 15:19 |
Action: lekernel trying tabula. 9GB download, flexlm DRM shitware, installer taking 33% CPU to download files from the internet, UI toolkit from 1991, installer crashing when mirror is unavailable. so far, sounds familiar. | 15:19 | |
Fallenou | ahah | 15:20 |
juliusb | I don't know what tabula is but I'll assume it's EDA from the description of the process you're going through | 15:20 |
lekernel | a new FPGA company that claims to achieve halfway decent (2GHz counters) performance | 15:21 |
lekernel | tabula.com | 15:21 |
wpwrak | milkymist as a project has more or less ended already, so why rename it now ? | 15:21 |
juliusb | ohh yes, these guys | 15:21 |
wpwrak | if you don't like the name anymore, give new things a new name .. | 15:21 |
Fallenou | 2 GHz, sounds attractive :) | 15:24 |
juliusb | holy hell, that ABAX A1EC04 thing has nearly 400k LUTs, and 5.5MB of block ram?!?! How much does it cost? | 15:24 |
juliusb | ahhh, yes, $7,500, not quite enthusiast gear | 15:24 |
lekernel | they have smaller and cheaper ones it seems | 15:25 |
lekernel | wpwrak, M1 is dormant, but the project (as a OSHW community) has not ended | 15:25 |
juliusb | the virtex7 2000T parts have about that much block ram but only 1/3 of the LUTs, but are an order of magnitude more expensive in development kit form | 15:26 |
lekernel | the 2000T has nearly 2 million LUTs | 15:27 |
lekernel | no? | 15:27 |
lekernel | but half the speed of the tabula stuff | 15:28 |
juliusb | 305k slices and each slice has 4LUTs and 8FF | 15:28 |
juliusb | it has nearly 2 million "logic cells" | 15:28 |
lekernel | mh | 15:28 |
lekernel | new confusing terminology from xilinx marketing ... | 15:28 |
juliusb | oh they always do this | 15:28 |
juliusb | but yeah probably those tabula things are faster, but, how do the same designs fare on both? Tablua are new, and it's unlikely their backend tool is as well tested as the Xilinx stuff | 15:29 |
wpwrak | lekernel: i mean that you probably won't make another vj station. that's what's associated with the name "milkymist". of course, there's a diverse community around it, but if it's for things like mailing lists and such, just creating a new places would be no more intrusive than a rename | 15:30 |
wpwrak | heh, see ? "new confusing terminology" :) | 15:30 |
lekernel | juliusb, given how buggy the xilinx stuff is, I can imagine that a skilled team of developers can do better | 15:31 |
lekernel | but I must say that so far I'm not impressed with the tabula software | 15:31 |
juliusb | what are you trying to put through it? | 15:32 |
lekernel | still at the bloated install process now... then I have to wait another 24 hours for the license approval... meh | 15:32 |
lekernel | but there are already familiar red flags in the installer: flexlm, 9GB download, outdated UI toolkit, etc. | 15:33 |
lekernel | and even the fact that there is an installer at all... | 15:34 |
juliusb | ah these FPGA guys love this shit | 15:39 |
wpwrak | the more shackles the better | 15:40 |
Action: Fallenou dreams of a 500 MHz lm32 core | 15:44 | |
juliusb | build an ASIC | 15:48 |
juliusb | and why are you dreaming of that lame core, mor1kx every day of the week man ;) | 15:49 |
lekernel | hmm, last time I checked that 'lame core' was still several times the speed and several times less area than mor1kx - and it has a MMU | 15:50 |
juliusb | mor1kx has MMU now too (stekern's great work) | 15:51 |
lekernel | idk why everyone in the open FPGA community wants to have their own processor, this results in a huge graveyard of unfinished, buggy and unusable projects | 15:51 |
juliusb | or1200 is not unfinished, buggy or unusuable, mor1kx is very much actively under development | 15:51 |
juliusb | or1200 is far from optimal though, and is the point of the mor1kx | 15:51 |
lekernel | M1 could not run on OR1200 due to low performance. that qualifies as 'unusable' for me. | 15:52 |
juliusb | but you're right, everyone does want their own ;) because we think we can do it better | 15:52 |
juliusb | lekernel: you're probably just using it wrong, it runs OK with caches | 15:52 |
lekernel | there were even latches inferred in the HDL when I tried it - I will admit that you have fixed that now, but still it was buggy | 15:52 |
juliusb | anyway, that's enough stirring from me :) | 15:52 |
lekernel | well in the end there's only one argument you can use | 15:53 |
juliusb | for what? | 15:53 |
lekernel | mibench (or equivalent) results | 15:53 |
lekernel | accompanied with a LUT count :-) | 15:53 |
juliusb | well, I think for FPGA stuff, synthesisable speed and coremark result (I presume mibench is similar?) | 15:54 |
juliusb | FPGAs are big enough now not to worry too much about LUT count | 15:54 |
lekernel | not really, I'm toying with the idea of a entry-level video mixer based on 6slx9 or even 6slx4 atm | 15:54 |
juliusb | I have heard the numbers of the lm32, it is impressive and I'm not sure why - could be architectural - or1k has some bad things in there | 15:55 |
juliusb | well you're a masochist then | 15:55 |
juliusb | on the other hand, you will have quick compile times :) | 15:56 |
lekernel | why? the current milkymist-ng design, including ethernet which I will remove, fits effortlessly in lx9 | 15:56 |
lekernel | yes, that too | 15:56 |
juliusb | seriously? which ethernet are yuo using? the Xilinx stuff or do you have your own? | 15:56 |
lekernel | own. I'm not using any Xilinx stuff. | 15:56 |
juliusb | ohh i need to look at that | 15:56 |
juliusb | the OpenCores one is massive and i tried to fix it but failed a long time ago | 15:57 |
juliusb | a better one would be nice | 15:57 |
lekernel | yes, I tried it | 15:57 |
lekernel | well mine is very minimal | 15:57 |
juliusb | that's fine | 15:57 |
juliusb | i gotta, go, ttyl | 15:57 |
lekernel | no DMA, just block RAM, and none of the stupid/legacy ethernet features implemented | 15:57 |
lekernel | works fine at slow speeds (due to CPU copying lots of memories, doing the CRCs, etc.) though | 15:58 |
Fallenou | farnell and digikey don't sell tabula devices :( | 17:39 |
--- Wed Jun 5 2013 | 00:00 |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!