Thanks for the interesting info!

My guess would be iHeartMedia

0x1E is the record separator, in ASCII precisely for this purpose. Too bad it’s not popular, here we’re stuck with inferior TSV/CSV

I can't easily type that out - and once the format can't be read / editing in a simple text editor, I'm starting to lean towards a nice binary format like protobuf.

Strings can contain 0x1E, so it has exactly the same issues as a tab character but with all the downsides of it not being an easy, “simple” character.

This is insanely fast, obviously a game changer over time. You should try the demo!

This seems to be using custom inference-only HW. It makes a ton of sense to use different HW for inference vs training, the requirements are different.

Nvidia, as far as I can tell, is focusing all-in on training and hoping the same HW will be used for inference.

Exciting times!

Hi there, I work for Groq. That's right. We love graphics processors for training but for inference our language processor (LPU) is by far the fastest and lowest latency. Feel free to ask me anything.

What's the scale of hardware behind this demo, in terms of watts, transistors and cost?

Are they only available on the cloud? Are you planning on releasing a consumer version?

Mostly available as a service via cloud API at the moment. The systems themselves are too big for consumers but we will sell systems to corporations.

The animations at the bottom are really fascinating. I wonder if it's some kind of simulated annealing algorithm(s)?

That is correct, VPR's placement is done using simulated annealing.

See section 8 of this paper (https://dl.acm.org/doi/pdf/10.1145/3388617), the originals VPR paper (https://link.springer.com/chapter/10.1007/3-540-63465-7_226), and the source code (https://github.com/verilog-to-routing/vtr-verilog-to-routing...) for more details.

Thanks for the info. Been using one of the knots for years, donated now.

I was debugging a Heisenbug once, developing embedded FW for a mobile phone.

After some time, I noticed that the phone seemingly only crashed in one area of the open office floorplan where I was working.

I started walking around the office testing this theory, not really believing it. But after a while, I had hard evidence that the bug would only manifest once I entered that part of the office.

When I came to terms that I wasn’t hallucinating, I realised what the problem was. There was poor reception in that part of the office, causing the phone’s modem to switch from 4G wideband to narrowband (glossing over details here), which triggered the bug.

Easy to see with hindsight, but I was very confused there and then

Here's a complementary tool for creating (scatter) plots of the CSV-file, note that this the 100k rows example from your link:

https://csvplot.com/remote_file.html?url=https://media.githu...

Thank you. I don’t know what I expected to hear, but it wasn’t that

In case the author sees this: Thank you for enabling CORS so that it's possible to plot examples from other sites. It would be awesome if the Content-Range header was allowed as well

Here is an example of plotting the first dataset that popped up for me: https://csvplot.com/remote_file.html?url=https://csvbase.com...

Wow that is cool, really cool.

I will look into implementing Content-Range - what is it that you want that for? What's the usecase?

Thank you for csvbase, today is the first time I've seen it.

I believe PapaParse, a JS library for parsing CSV files, uses Content-Range to stream large CSV files in chunks.

https://csvplot.com uses PapaParse under the hood, I saw a warning in the dev console and posted here. I'm not sure why it seemingly works fine anyway.

You are most welcome. So glad it is useful.

This subject tracked here: https://github.com/calpaterson/csvbase/issues/29

Magic!