@technology@lemmy.world Did startup Flow Computing just make CPUs 100x faster?
https://www.theverge.com/2024/6/11/24176304/flow-computing-startup-parallel-processing-accelerator
copying from another comment
Betteridge’s law of headlines is an adage that states: “Any headline that ends in a question mark can be answered by the word no.” It is based on the assumption that if the publishers were confident that the answer was yes, they would have presented it as an assertion; by presenting it as a question, they are not accountable for whether it is correct or not. The adage does not apply to questions that are more open-ended than strict yes–no questions.
Don’t you not want to send me the balance of your bank account? No means yes and yes means yes.
Did startup Flow Computing just make CPUs 100x faster?
No.
It’s probably not a total lie, a dedicated chip for specific calculations is probably the only way that we’ll see major jumps in processing power for a long while. It can be really effective for highly specific stuff. But the headline itself is certainly pure clickbait on the Verge’s and Flow’s behalf.
@antler@feddit.rocks Yeah, for sure, I thought the white paper and the idea was neat, and maybe even foreshadowing for what’s to come.
Probably should’ve put a more descriptive title, but I was just being lazy.
Looks like they’re just adding separate CPU cores, similar to how a GPU adds a different type of compute. I guess it can work, but why not just build that into the CPU to begin with?
Then again, I haven’t read the whitepaper, but I highly doubt there’s anything groundbreaking going on here…
You know, the answer to captions like that is to 99.9%:
Yes*
*Under laboratory conditions and for a very specific use case / a whole lot of money, once.
The reality is that billions are poured into developing faster computers and change is happening gradually, because low-hanging fruits are gathered even before they are ripe.
You can’t exceed lightspeed. Current tech is already at 99%
