|
What ever happened to this, lol. https://www.youtube.com/watch?v=pq79h-HA5lU https://www.youtube.com/watch?v=h70l8Hhe5c8 I forget if it was supposed to reinvigorate the anroid hardware market or if everyone thought it would flop out of the gate. so much for android being platform agnostic here's something else MIPS related if you want https://www.youtube.com/watch?v=1n0kWWqQ0D8
|
#
?
Sep 10, 2017 03:01
|
|
|
|
| # ? Apr 24, 2024 07:22 |
|
|
i dunno but i just ate a mint chocolate frozen greek yogurt sandwich
|
|
#
?
Sep 10, 2017 03:02
|
|
|
go on...
|
|
#
?
Sep 10, 2017 03:03
|
|
|
my bike helmet has MIPS i don't like it
|
|
#
?
Sep 10, 2017 05:44
|
|
|
mips was a good architectural concept but some of the fundamental assumptions that it was based on are no longer true.
|
|
#
?
Sep 10, 2017 07:08
|
|
|
whats the point
|
|
#
?
Sep 10, 2017 08:25
|
|
|
The Management posted:mips was a good architectural concept but some of the fundamental assumptions that it was based on are no longer true. would love to hear this elaborated on
|
|
#
?
Sep 10, 2017 10:22
|
|
|
pram posted:whats the point well, mips derivatives were used in the playstation, ps2 and nintendo 64, so, video games
|
|
#
?
Sep 10, 2017 16:45
|
|
|
i'm running android on my n64
|
|
#
?
Sep 10, 2017 19:48
|
|
|
anrouide
|
|
#
?
Sep 10, 2017 20:44
|
|
|
here's some mips for you https://youtu.be/nUevMxh5od0
|
|
#
?
Sep 11, 2017 01:53
|
|
|
echinopsis posted:would love to hear this elaborated on
|
|
#
?
Sep 11, 2017 02:00
|
|
|
MIcroPeniS
|
|
#
?
Sep 11, 2017 02:03
|
|
|
eschaton posted:here's some mips for you drat androids are bulky and loud
|
|
#
?
Sep 11, 2017 04:16
|
|
|
Hed posted:MIcroPeniS lol
|
|
#
?
Sep 11, 2017 04:40
|
|
|
drat why did i risc logging back in today
|
|
#
?
Sep 11, 2017 06:08
|
|
|
hello
|
|
#
?
Sep 11, 2017 07:40
|
|
|
hi
|
|
#
?
Sep 11, 2017 10:26
|
|
|
yo
|
|
#
?
Sep 11, 2017 16:09
|
|
|
Tom Collins posted:drat why did i risc logging back in today were you strongARMed?
|
|
#
?
Sep 11, 2017 16:55
|
|
|
Tom Collins posted:drat why did i risc logging back in today wow, tarm collins still visits SA? noice
|
|
#
?
Sep 12, 2017 02:19
|
|
|
why wouldn't you want the power of SGI, PS1 & PS2, and N64 in your anroid?
|
|
#
?
Sep 12, 2017 02:20
|
|
|
I'm looking forward to the 20 year retro
|
|
#
?
Sep 12, 2017 07:52
|
|
|
Doc Block posted:why wouldn't you want the power of SGI, PS1 & PS2, and N64 in your anroid? sounds like a good deal!
|
|
#
?
Sep 12, 2017 15:11
|
|
|
The Management posted:mips was a good architectural concept but some of the fundamental assumptions that it was based on are no longer true. still want to hear more about this
|
|
#
?
Sep 12, 2017 20:11
|
|
|
a big part of what made RISC in general come about in the 80s was that RAM was really fast relative to the CPU. so the idea with RISC was that you have a CPU that only has a small, simple ISA with no complex addressing and no microcode, so the CPU is simpler but has to execute more instructions (thus larger code size) to accomplish the same task. the idea being that being simpler meant you could make it faster, the bigger code size and generally having to touch memory more often wouldn’t matter because memory was so fast, and having to execute more instructions wouldn’t matter because each instruction would take less work, giving a net speed increase. memory is waaaay slower than the CPU now. and memory being way slower than the CPU was what gave rise to CISC in the first place, back before RISC was a thing. Doc Block fucked around with this message at 00:02 on Sep 13, 2017 |
|
#
?
Sep 12, 2017 23:13
|
|
|
cool that's interesting i like
|
|
#
?
Sep 13, 2017 02:30
|
|
|
Doc Block posted:a big part of what made RISC in general come about in the 80s was that RAM was really fast relative to the CPU. so the idea with RISC was that you have a CPU that only has a small, simple ISA with no complex addressing and no microcode, so the CPU is simpler but has to execute more instructions (thus larger code size) to accomplish the same task. the idea being that being simpler meant you could make it faster, the bigger code size and generally having to touch memory more often wouldn’t matter because memory was so fast, and having to execute more instructions wouldn’t matter because each instruction would take less work, giving a net speed increase. neat post ty
|
|
#
?
Sep 13, 2017 02:32
|
|
|
except RAM wasn't very fast relative to the CPU even by the mid-late 1980s while (say) an 8MHz 68000 gave you 125ns bus cycles, workstation class hardware was going to 25MHz with a corresponding 40ns cycle time by 1987-88, hence the substantial growth of L2 and then L1 cache hell, my circa 1989 Mac IIci had a slot for a cache RAM card, I had the Apple one that added 32KB of 35ns cache and made a visible difference, and Micron made up to a 128KB cache card RISC was proposed as a win because instead of complex operations each taking multiple cycles, you could do simple operations at an average of one per cycle (given good instruction pipelining and branch prediction, and compiler assistance with that) and wind up with higher throughput, especially by isolating external memory access (even cache) from other processing it actually was a win in that way too, it certainly beat the CISC designs of the day through the early 2000s, many people just didn't foresee Intel (and AMD) being able to keep up by implementing dynamic instruction translation in hardware like they wound up doing of course once they got the low hanging fruit, the RISC designs started adding more complicated instructions that were harder to implement and make faster with chip generations, to the point where PowerPC was mocked at its introduction for having more instructions than the “CISC” 6502, which culminated with everyone adding Cray-style vector instructions
|
|
#
?
Sep 13, 2017 02:55
|
|
|
one thing I want to see is the relative performance of something like the BYTE UNIX benchmark on a wicked awesome late 1990s SGI workstation and a little embedded Linux based router or something with a MIPS SoC
|
|
#
?
Sep 13, 2017 03:00
|
|
|
i bow to your superior knowledge. i thought the "RAM as fast as the CPU" thing stayed true up until the late-ish 80s.
|
|
#
?
Sep 13, 2017 03:50
|
|
|
I think the real question here is, why is ram so slow
|
|
#
?
Sep 13, 2017 03:56
|
|
|
when the ram is slow <<<<<<<<<<<<<<<<<<<<
|
|
#
?
Sep 13, 2017 05:21
|
|
|
can't wait for DDR-X
|
|
#
?
Sep 13, 2017 07:27
|
|
|
Also, ARM is a RISC and (32 bit) ARM also has 80s assumptions baked into it (like making the program counter just a regular register - yes, you can technically execute a jump by mov pc, <some reg>) MIPS didn't lose out in the embedded space because it sucks; it's more like how Google Plus didn't pan out because Facebook already ate their lunch. ARM is already absolutely everywhere and everyone knows how to work with it and loads of people make products for it, so what's MIPS got that ARM doesn't, other than being able to cheap out on licence fees (which is nice, yes, but not nice enough to counteract all the other stuff). Everyone else uses ARM, so why wouldn't you? (this is also, btw, why Android on x86 never really took off) Imagination pushed MIPS because they desperately wanted there to be an alternative to ARM that they controlled and because if they managed that they wouldn't be reliant on selling PowerVR to Apple for revenue and now welp Apple's making its own GPU and they're going to be broke now and they're trying to flog off their MIPS business. feedmegin fucked around with this message at 16:25 on Sep 13, 2017 |
|
#
?
Sep 13, 2017 16:22
|
|
|
echinopsis posted:can't wait for DDR-X rambus 
|
|
#
?
Sep 13, 2017 16:31
|
|
|
optane
|
|
#
?
Sep 13, 2017 16:58
|
|
|
Clicked thinking thread was called "I'm going to enjoy my ice cream sandwich, with MPLS!"
|
|
#
?
Sep 13, 2017 21:53
|
|
|
RISCy Business posted:optane now optane is something I can get into
|
|
#
?
Sep 14, 2017 07:00
|
|
|
|
| # ? Apr 24, 2024 07:22 |
|
|
Schadenboner posted:Clicked thinking thread was called "I'm going to enjoy my ice cream sandwich, with MPLS!" it's a good city
|
|
#
?
Sep 14, 2017 20:54
|
|