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

I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

news.ycombinator.com

Post reply on HN
Print view View original

111–120 of 129 posts

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#111

Post by whattheheckheck »

satvikpendem wrote: post #69
goodmythical wrote: post #64

Earlier quoted context omitted.

You don't mind Google using your kids to train their models and advertising algorithms? Years from now they'll be getting "hey look at BIKE BRANDS' NEWEST CHEAP BIKE REMEMBER WHEN YOU USED TO RIDE BIKE BRAND BIKES"

I think most people really don't care, and/or will just adblock those sorts of things when they do arrive.

What about in 10 years when they auto search and label users for political dissent and likelihood of impact

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#113

Post by satvikpendem »

whattheheckheck wrote: post #111
satvikpendem wrote: post #69

Earlier quoted context omitted.

I think most people really don't care, and/or will just adblock those sorts of things when they do arrive.

What about in 10 years when they auto search and label users for political dissent and likelihood of impact

That's for future them to worry about, they'll think. Honestly I've never seen people historically care about privacy even if it means the government rounds them up, people simply don't seem to care at all, and trying to make them care is a futile effort in my experience.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#114

Post by voidmain0001 »

Rohansi wrote: post #75
voidmain0001 wrote: post #68

Earlier quoted context omitted.

No comparison. M1 Max has 400GB/s RAM bandwidth while Snapdragon X2 Elite, the latest and greatest , has 228GB/s RAM bandwidth.

I don't disagree with your conclusion but the comparison of max bandwidth between the two SoCs is not enough. Neither of them will use all of that bandwidth doing AI work because the GPU will be compute limited. That's why dedicated GPUs perform so significantly better without having significantly higher bandwidth.

The question I answered was "Would Windows ARM performance be similar for those cpu?" and the answer is, no, because the RAM bandwidth for ARM SoC computers for Windows, primarily Snapdragon X1E and X2E is half to quarter that of the M1 Max.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#115

Post by crakhamster01 »

Thanks for sharing! I make videos and often have the same problem as well.

Being able to semantic search over your library is useful, but does it solve the review problem? I feel like you would still need to watch the footage back before you know what you're working with.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#116

Post by iliashad »

crakhamster01 wrote: post #115

Thanks for sharing! I make videos and often have the same problem as well. Being able to semantic search over your library is useful, but does it solve the review problem? I feel like you would still need to watch the footage back before you know what you're working with.

You're welcome. As of now, the project narrow down your options instead of watching 3-4 videos (5 min each, as an example), you can watch 3 clips from them instead (1s to 2.5s each). It gives you only the video moments that you need to watch for review instead of watching the full videos.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#117

Post by iliashad »

Schiendelman wrote: post #92
robrain wrote: post #35

Earlier quoted context omitted.

Haven’t tried it yet, and I don’t know if it matches OP’s requirements, but the blurb says “You can even search for individual faces” https://www.blackmagicdesign.com/products/davinciresolve/wha...

This is what took me from free to paid user, and it was well worth it.

That's good, also the Davinci resolve integration works with the Free and Studio version the same way.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#118

Post by Schiendelman »

iliashad wrote: post #117
Schiendelman wrote: post #92

Earlier quoted context omitted.

This is what took me from free to paid user, and it was well worth it.

That's good, also the Davinci resolve integration works with the Free and Studio version the same way.

Oh yeah - I just wanted to support them!

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#119

Post by Rohansi »

voidmain0001 wrote: post #114
Rohansi wrote: post #75

Earlier quoted context omitted.

I don't disagree with your conclusion but the comparison of max bandwidth between the two SoCs is not enough. Neither of them will use all of that bandwidth doing AI work because the GPU will be compute limited. That's why dedicated GPUs perform so significantly better without having significantly higher bandwidth.

The question I answered was "Would Windows ARM performance be similar for those cpu?" and the answer is, no, because the RAM bandwidth for ARM SoC computers for Windows, primarily Snapdragon X1E and X2E is half to quarter that of the M1 Max.

Same thing applies. Neither CPU would be able to fully saturate the available memory bandwidth so comparing by bandwidth alone is not accurate.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#120

Post by duncangh »

Would it be possible to upload them all to YouTube as unlisted and then point Gemini to them? Not sure the limitations wrt unlisted videos. Maybe also could’ve been done with the @google photos operator if uploaded to Google Photos if not a corporate google workspace Gemini instance
Post reply on HN