2 ms·

I'm not convinced this is true and not just a popular argument. Just today I've processed 4 TB of JSON data, in Python, on a desktop computer. That would be cr

I'm not convinced this is true and not just a popular argument. Just today I've processed 4 TB of JSON data, in Python, on a desktop computer. That would be crazy to imagine 10 years ago. Yes, many cores, fast SSD, etc. Maybe it should be 20 TB if ultra optimized, but the capability advance is incredible. Yes, apps maybe start slowly, but now everybody can easily edit HD video on their mid-range laptop.

grumpyprole 3y ago
A complied language can be over 1000 times faster than Python, so maybe the dual core from 10 years ago would have been fine.
- surfingdino 3y ago
  Yes. I am writing more code in Golang for that reason. I'll take the 100x increase any day.
xcv123 3y ago
> That would be crazy to imagine 10 years ago We had this in 2013. Would have handled the job with a well written parser. https://www.intel.com/content/www/us/en/products/sku/77779/intel-core-i74960x-processor-extreme-edition-15m-cache-up-to-4-00-ghz/specifications.html https://www.intel.com/content/www/us/en/products/sku/77779/i...
wharvle 3y ago
"Computers are in fact very fast" and "certain programs do in fact demonstrate that" and "the real-world experience of using computers for most people is that performance is largely treading water or going backwards" due to a boom in software inefficiency, can all be true at the same time.
anthk 3y ago
No, not crazy: https://adamdrake.com/command-line-tools-can-be-235x-faster-than-your-hadoop-cluster.html https://adamdrake.com/command-line-tools-can-be-235x-faster-... ~10 years ago I parsed 15GB of Windows binary logs with xargs, grep and strings(1) in minutes. Minutes. Less than 1/4 of an hour.