Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

3.14.by

Post reply on HN
Print view View original

141–150 of 158 posts

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#141

Post by brucehoult »

camel-cdr wrote: post #68
alias_neo wrote: post #61

Earlier quoted context omitted.

> Do not forget to preorder, to get the unbelievably low price listed in their announcement. I don't know much about the RISC-V ecosystem at the moment, but what do you consider an "unbelievably low price"? They SoC appears to be £5 to pre-order (EDIT: £5 is for a 20% off coupon), but the motherboard to use it on appears to be ~£1200. Is this considered a good deal for RISC-V right now?

The unbelievably low price price milk-v listed was $120. Sipeed mentioned something abour $300. I presume that the milk-v one might be just the cpu+motherboard, and sipeed included ram and gpu in the price, but thats all speculation. If it's anywhere in the $300 range I'd be very happy. Just for scale the specint numbers from the P670 are on Cortex A76 level, so you get what the rp5 has, but you get way more cores.

> Just for scale the specint numbers from the P670 are on Cortex A76 level,

A78 level, 12 SPECint2006/GHz.

It's P470 (Horse Creek, if it ever appears) and Dubhe-90 that are at A76 8 SPECInt2006/GHz level.

And, for completeness, P870 is 18 SPECInt2006/GHz.

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#142

Post by brucehoult »

adgjlsfhk1 wrote: post #113
FullyFunctional wrote: post #19

Earlier quoted context omitted.

Completely false. This is compared to native versions on all platforms. Supposedly TH1520 gets a small uplift when compiled for its proprietary extensions, but I haven't seen a compiler using this and it would have no bearing on application from a standard distribution. I have both 7110 and TH1520, but I'm looking forward to SiFive P670 powered SG2380 which promises to be an exciting chip.

it is a problem that the compilers for riscv are still pretty bad. it will take a few years (and more cheap risc-v dev boards) before gcc and clang actually know how to generate good code for risc-v (this is especially a problem for things like video payback and encryption that tend to be written with significant amounts of hand tuned assembly)

There is no problem with the compilers for RISC-V, what are you talking about?

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#143

Post by brucehoult »

999900000999 wrote: post #97

I'd suggest anyone who wants to play with RISC-V to buy a MangoPi instead. The software still isn't really in a working state. But I'd rather pay 30$ to figure that out over 800$.

Of course a laptop with case, screen, battery etc costs more than a bare SBC.

The Milk-V Mars, with very similar performance to this (see other comments for JH7110 vs TH1520), starts from $39 with 2 GB RAM (and 4 GB or 8 GB options) and of course quad core 1.5 GHz dual-issue CPU. That's around 8x the compute performance of the 1 GB MangoPi MQ-pro, for $10 more. Or, for more bare-bones, the Mars CM is $34.

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#144

Post by adgjlsfhk1 »

brucehoult wrote: post #142
adgjlsfhk1 wrote: post #113

Earlier quoted context omitted.

it is a problem that the compilers for riscv are still pretty bad. it will take a few years (and more cheap risc-v dev boards) before gcc and clang actually know how to generate good code for risc-v (this is especially a problem for things like video payback and encryption that tend to be written with significant amounts of hand tuned assembly)

There is no problem with the compilers for RISC-V, what are you talking about?

the compilers work (produce valid code), but things like chip specific instruction cost models, full support for vector operations, and stuff like that are nowhere near where they are for x86 or M1 (because there are way less users and developers to make sure the codegen is optimal)

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#145

Post by theanonymousone »

saxonww wrote: post #85
sylware wrote: post #63

Earlier quoted context omitted.

RISC-V is becoming a worldwide interop standard, and a real one, namely royalty free and which will be stable in time once we get all the "now we know" features, and all that at the ISA level (x86_64 and ARM are _NOT_ worldwide royalty free). And don't be mistaken, there is no perfect ISA (it is said "average"), it does not exist, only a set of trade-offs and compromises which will not fit all use-cases perfectly. RI…

This is the first time I've ever seen anyone write 'silicium' and sure enough, it's an alternative name for silicon. I wonder if this is a more common name for silicon in some other part of the world (I'm in the US). I don't think I've ever seen this spelling before today, and I'm not a young person.

Yes. We knew it as silicium in the school and it took me some time to realise that silicon is the same thing.

For a long time I assumed that Silicon is the "unofficial", American word for it, while silicium is the scientific name.

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#146

Post by brucehoult »

adgjlsfhk1 wrote: post #144
brucehoult wrote: post #142

Earlier quoted context omitted.

There is no problem with the compilers for RISC-V, what are you talking about?

the compilers work (produce valid code), but things like chip specific instruction cost models, full support for vector operations, and stuff like that are nowhere near where they are for x86 or M1 (because there are way less users and developers to make sure the codegen is optimal)

Core-specific cost models are a bad idea when you have a large variety of cores in use. Auto-vectorisation doesn't work well on anything, except in the most trivial cases, and none of the popular boards [1] yet have the official RISC-V vector ISA anyway -- that will change this year, hopefully.

[1] there is now in the last month or two one (1) board with RVV 1.0 but it's only single core vs the quad core in popular boards, and also has only 512 MB of RAM vs 4 / 8 / 16 GB on the popular boards (and 128 GB on the Pioneer).

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#147

Post by FullyFunctional »

adgjlsfhk1 wrote: post #144
brucehoult wrote: post #142

Earlier quoted context omitted.

There is no problem with the compilers for RISC-V, what are you talking about?

the compilers work (produce valid code), but things like chip specific instruction cost models, full support for vector operations, and stuff like that are nowhere near where they are for x86 or M1 (because there are way less users and developers to make sure the codegen is optimal)

[citation needed] Do you have anything concrete to support "nowhere near where they are for x86 or M1"? I literally eyeball RISC-V code all day long for a living and it's not like there a massive performance lost to the compiler.

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#148

Post by snvzz »

Havoc wrote: post #81

How difficult is working with risc v currently? I’m guessing back Linux works but most precompiled stuff not?

Debian officially supports RISC-V and is about there[0].

Whereas qemu-user will run e.g. x86 and arm binaries, should you ever need to.

This recent review of VisionFive2[1] has a little more info. There they even run popular games via x86 emulation.

0. https://buildd.debian.org/stats/

1. https://boilingsteam.com/vision-five-2-board-review/

Re: Lichee Console 4A – RISC-V mini laptop: Review, benchmarks and early issues

#150

Post by sylware »

hajile wrote: post #125
sylware wrote: post #63

Earlier quoted context omitted.

RISC-V is becoming a worldwide interop standard, and a real one, namely royalty free and which will be stable in time once we get all the "now we know" features, and all that at the ISA level (x86_64 and ARM are _NOT_ worldwide royalty free). And don't be mistaken, there is no perfect ISA (it is said "average"), it does not exist, only a set of trade-offs and compromises which will not fit all use-cases perfectly. RI…

RISC-V is already putting the squeeze on ARM in the embedded market. RISC-V is fewer gates at almost every performance level the embedded market is interested in due to having a tighter, non-legacy ISA (there's even the open SERV core which fits RISC-V into an insanely small 2.1KGe). RISC-V likely has a future in mobile devices as it crawls up the performance tree like ARM did. This leaves ARM in a tight squeeze betw…

I am waiting for this RISC-V Cortex 42... they better move there, and everybody will be happy.
Post reply on HN