naveen_k

I have been experimenting with using 5 phases of movements with each phase covering different areas of the screen while being actively moving. The last phase makes the dot move in a Lissajous-like motion which is more fluid like you are suggesting.

The challenge is recording and syncing the motion at a higher frequency and being able to save without much drift and the performance of these landmark/gaze models is often slow.

One more option to speed it up is not to do the eye tracking at record time, just record a crop video of the face and the screen first at 60Hz and then run the model on each frame and update the metadata of the dataset.

Ha! The timing is impeccable. This is a great demo. I've been experimenting with using gaze and eye tracking for cursor prediction as a side project. I like the idea of pressing 'space' for each dot. I just had the 9-dot movement going from one point to another. I'm using Mediapipe's face landmarks model (I wasn't aware of WebGazer). I'll continue to work on it, but it's great to see a similar thought process.

That's exactly what I did first! I ran a CPU torture test at full clock speed and measured the power draw at the plug, then repeated the same test with the lowest clock speed setting. For the Epyc system, there was about 225W lower power draw at the reduced clock speed. Even at idle, capping the max frequency reduced the power draw by about 20+%.

Thanks! I initially just wanted to build a dashboard, with the power optimization part being a later addition. Based on the HN response, it seems that's the feature that resonated most with people. I'll be making improvements to the optimization component in the coming days and will publish what I have.

That's actually pretty cool. ESPs are awesome little things.

It's a 1600W PSU (Coolmaster 1600 V2 Platinum)

I'm actually using a 1600W PSU. 1400W is my target max draw. This is a dual EPYC (64 core CPU each) system btw. The max draw by the CPU+MB+Drives running at peak 3700MHz without the GPU is 495W! Adding 4x 4090 (underclocked) will quickly get you to 1400W+.

Thanks, I'll check it out.

Quick update: Definitely wasn't expecting this to end up on the front page. I was more focused on publishing the dashboard than the power optimizer service I'm running. I'll take all the feedback into account and will open source an improved version of it soon. Appreciate all the comments!

Thanks for the suggestion! I'm actually using dual EPYC server processors in this workstation, not Ryzen. I'm not sure EPYC supports PBO/Curve Optimizer functionality that's available in AM4/AM5 platforms.

That said, I'm definitely interested in learning more about processor-specific optimizations for EPYC. If there are server-focused equivalents to what you've mentioned that would work better than frequency capping, I'd love to explore them!

Thanks for the suggestion. Will check it out.

Sorry, I only open sourced the dashboard part as mentioned in the bottom of the blogpost. Still working on improving the 'Power optimizer' service so will open source that soon as well.

Thanks! That's an excellent point. You're right that there's likely a sweet spot that would be more efficient overall than aggressive throttling.

The current implementation uniformly sets max frequency for all 128 cores, but I'm working on per-core frequency control that would allow much more granular optimization. I'll definitely measure aggregate consumption with your suggestion versus my current implementation to see the difference.

Good point. I'm often running multiple parallel jobs with varying priorities where uniform throttling actually makes sense. Many LLM inference tasks are long-running but not fully utilizing hardware (often waiting on I/O or running at partial capacity)

The dual Epyc CPUs (128 cores) in my setup have a relatively high idle power draw compared to consumer chips. Even when "idle" they're consuming significant power maintaining all those cores and I/O capabilities. By implementing uniform throttling when utilization is low, the automation actually reduces the baseline power consumption by a decent amount without much performance hit.

The power optimizer daemon has only been running for a few days, so it's hard to measure in $ value but based on my peak pricing I would estimate the savings to be around a few dollars since then.

Definitely, I've been contemplating getting a 5-10kWh LFP battery backup with <10ms UPS switchover to run the workstation and home backup. This is an intermediate solution until then.

A cronjob would definitely work in most cases if the goal is just to auto change freq profiles during set ToU periods. I just wanted a more flexible system where the system can auto change the profiles based on actual utilization so demanding tasks aren't slowed down.

I've been putting off starting a blog to document my side projects for a long time. Finally having it up feels great - it motivated me to organize my thoughts and publish rather than letting them sit as unstructured notes.

Thanks!

For audio based detection, I would start with collecting a large corpus of normal shows vs ad segment audio dataset. I assume the amplitude (volume changes) and frequency distribution would be unique enough to distinguish between them.

Train a model using their Mel Spectrogram and deploy on-device.

Microphone -> ADC -> Preprocessing -> Spectrogram Generation -> Inference -> Mute/Unmute

Would be an interesting project.

FYI, the project it posted here. https://news.ycombinator.com/item?id=42856980

Perfect timing! I got annoyed by Ads in streaming services and decided to build something to mute the TV when an Ad comes on. Simple vision based system.

I'll post a write up in a few days.

This looks exciting! Will definitely be testing it out in the coming days.

I see you offer re-ranking using local models, will there be build-in support for making re-ranking calls to external services such as cohere in the future?

Try changing the number of icons (Customize Top Level Menu) on the AssistiveTouch to anything other than the default six before your next os update. It’ll reset to the default layout.

If you use AssistiveTouch on your iPhone or iPad with a customized top-level menu, every iOS/iPadOS update resets it to default settings. Apple hasn't fixed it for over 8+ years either.

Not sure if anyone noticed but their intro video uses a voice-clone of Sam Altman. https://www.goody2.ai/video/goody2-169.mp4