31 ms·

I've always had a 20% threshold If anyone is promising a 20% or greater improvement over the current leaders, I immediately file it away as a scam until I see

I've always had a 20% threshold If anyone is promising a 20% or greater improvement over the current leaders, I immediately file it away as a scam until I see extraordinary proof. Other than some very immature industries, even 20% leaps don't happen without earth shattering once-in-a-generation breakthroughs. Some call me a pessimist, and occasionally I'm unfairly dismissive of new technologies. But for every one I get wrong, I get 99 others correct.

randomsearch 8y ago
When Google first arrived, I’d say the improvement was way more than 20%. Did you experience that?
- Rexxar 8y ago
  Search was not a mature industry at all.
  - randomsearch 8y ago
    Compared to AI? I wouldn’t say search was more immature than Cloud APIs for AI, if that comparison even makes sense.
  - ethbro 8y ago
    I'd definitely include that as a qualifier to the rule. Mature and/or competitive. 20%+ without a breakthrough can absolutely happen in niche industries that have historically lacked competition.
smnrchrds 8y ago
How do you define immature? AI seems to be quite immature even compared to other subjects in computer science, which is as a whole quite young.
scottlegrand2 8y ago
CUDA delivered a 10x to 100x performance boost on legacy code for real. The hardware had been capable of doing this for at least a generations previously but it lacked a few operations to make it accessible and efficient. What I think made people miss it is no one wanted to refactor their code to run efficiently on GPUs. It wasn't that hard to do so in a lot of cases but it was work. And all this set the stage for deep learning. It was the opening act if you will...
- sk5t 8y ago
  > CUDA delivered a 10x to 100x performance boost on legacy code CUDA offered a way to access a limited set of operations on special-purpose hardware with a rewrite of critical sections. A pretty far cry from 'boosting legacy code.'
megy 8y ago
> If anyone is promising a 20% or greater improvement over the current leaders, I immediately file it away as a scam until I see extraordinary proof. Ha, this is just silly. Ok, so what percent were IBM promising? How do you ever reduce anything complicated into a percent change.
Bjartr 8y ago
I feel like people are interpreting your "20%" criteria more literally than you intended.

randomsearch 8y ago

When Google first arrived, I’d say the improvement was way more than 20%. Did you experience that?

Rexxar 8y ago
Search was not a mature industry at all.
- randomsearch 8y ago
  Compared to AI? I wouldn’t say search was more immature than Cloud APIs for AI, if that comparison even makes sense.
- ethbro 8y ago
  I'd definitely include that as a qualifier to the rule. Mature and/or competitive. 20%+ without a breakthrough can absolutely happen in niche industries that have historically lacked competition.

smnrchrds 8y ago

How do you define immature? AI seems to be quite immature even compared to other subjects in computer science, which is as a whole quite young.

scottlegrand2 8y ago

CUDA delivered a 10x to 100x performance boost on legacy code for real. The hardware had been capable of doing this for at least a generations previously but it lacked a few operations to make it accessible and efficient. What I think made people miss it is no one wanted to refactor their code to run efficiently on GPUs. It wasn't that hard to do so in a lot of cases but it was work. And all this set the stage for deep learning. It was the opening act if you will...

sk5t 8y ago
> CUDA delivered a 10x to 100x performance boost on legacy code CUDA offered a way to access a limited set of operations on special-purpose hardware with a rewrite of critical sections. A pretty far cry from 'boosting legacy code.'

megy 8y ago

> If anyone is promising a 20% or greater improvement over the current leaders, I immediately file it away as a scam until I see extraordinary proof. Ha, this is just silly. Ok, so what percent were IBM promising? How do you ever reduce anything complicated into a percent change.

Bjartr 8y ago

I feel like people are interpreting your "20%" criteria more literally than you intended.