Hacker News Reader

byefruiton Lenovo ThinkPad P1 Gen3 Review – with Linux

If you scroll down on the link above you can see some graphs that show the 1280P is significantly faster than the M1 and the battery lives are very similar 11 hours vs 13 hours.

AMD are also doing a good job. The 7000s look very good. Crazy that this is still on a larger process node than Apple.

byefruiton Lenovo ThinkPad P1 Gen3 Review – with Linux

For what it's worth this is comparing against a 10th gen Intel. The 12th gen ones look like they'll outperform the M1s on a similar battery life:

https://www.notebookcheck.net/Dell-XPS-13-Plus-with-i7-1280P...

Note that that is a 12th gen P. I suspect the U CPUs won't perform quite as well but will offer much better battery life than the MBP.

byefruiton Framework Laptop Mainboard

My current laptop needs to be replaced really soon, I'm hoping Framework will announce either a recent AMD Ryzen or Intel's Alder Lake. The difference between the 11th and 12th gen Intel is big enough I don't want to spend the next three years regretting it.

Unfortunately Lenovo or Dell might get my budget if it looks like that's not going to happen for a while. This laptop has had a few too many drops off of my bike (which is why repairability would be great!).

byefruiton Update on the Atlassian outage affecting some customers

Having used Asana in several organisations, I think it being a suitable replacement for JIRA says more about how poor a choice JIRA was for situation than anything about Asana.

byefruiton The worst part of working from home is now haunting reopened offices

I do the 10-15 minute cycle and it is excellent. Have been considering throwing in the occasional walking day too.

byefruiton Bringing the Framework Laptop to more of the world

Intel only at the moment is correct but Windows only? TFA says "no OS" as an option, so you're free to install what you like.

byefruiton Playing with Matplotlib style sheets

Did not realise matplotlib had style sheets, that could have been useful. Thanks.

Also - unnecessarily gendered title?

byefruiton Intel’s Arc GPUs will compete with GeForce and Radeon in early 2022

That makes a lot more sense, thanks. They could do with making that a lot clearer on the project.

Still handicaps them compared to Nvidia where you can just buy anything recent and expect it to work. Suspect it also means they get virtually no open source contributions from the community because nobody can run or test it on personal hardware.

byefruiton Intel’s Arc GPUs will compete with GeForce and Radeon in early 2022

Unfortunately that support is via ROCm, which doesn't support the last three generations (!) of AMD hardware: https://github.com/ROCm/ROCm.github.io/blob/master/hardware....

byefruiton Intel’s Arc GPUs will compete with GeForce and Radeon in early 2022

I really hope this breaks Nvidia's stranglehold on deep learning. Some competition would hopefully bring down prices at the compute high-end.

AMD don't seem to even be trying on the software-side at the moment. ROCm is a mess.

byefruiton Sam Stephenson Quits Basecamp and Related OSS Projects

Yea, there's a huge list of resignations:

https://twitter.com/sstephenson/status/1388146129284603906

https://twitter.com/ThatDetroitAndy/status/13881952733364019...

https://twitter.com/georgeclaghorn/status/138813100953171968...

https://twitter.com/lexicola/status/1388189598367559688

https://twitter.com/zachwaugh/status/1388190748189802501

https://twitter.com/sophaskins/status/1388155262972874754

https://twitter.com/jonasdowney/status/1388205182916440070

https://twitter.com/Rahsfan/status/1388209146487623681

https://twitter.com/kikiaards/status/1388208900608954371

https://twitter.com/mackesque/status/1388206605506842627

https://twitter.com/wcmoline/status/1388208323908968449

https://twitter.com/RossRochford/status/1388202323718115329

https://twitter.com/conormuirhead/status/1388207801646780416

13 resignations today by my count.

byefruiton AWS Lambda Edge changes duration billing granularity from 50ms down to 1ms

https://aws.amazon.com/blogs/aws/new-for-aws-lambda-1ms-bill...

Been 1ms for a few months now. =)

byefruiton Streamlit's $35M Series B

Looks interesting but has anyone managed to find pricing anywhere?

byefruiton We bootstrapped to $11M in ARR

Super interesting. Thank you.

byefruiton We bootstrapped to $11M in ARR

How did you ensure you'd given each iteration a proper go? Are there any tips you can share for validating them that were effective?

byefruiton MacBook Pro M1X benchmarks have leaked

Do you have any evidence for this claim? The actual concrete performance numbers for the M1 and Zen3 don't agree with you at all:

M1: https://images.anandtech.com/doci/16252/latency-m1_575px.png

Zen3: https://images.anandtech.com/doci/16214/lat-5950.png

If anything the M1's ram has _higher_ latency than the Zen3.

(Can't reply to the sibling comment but Intel's numbers are even better: https://images.anandtech.com/doci/16214/lat-10900.png)

byefruiton Launch HN: Albedo (YC W21) – Highest resolution satellite imagery

This looks really cool and congrats on the launch.

A few questions.

How big do you foresee your constellation being?

How fair is the comparison in your Google Photos link? The image on the right seems to have significant artifacting that looks more a side effect of compression.

byefruiton Replicating GPT-2 at Home

Alas, only if you live in the US.

Colab Pro isn't available outside the US (without breaking Google's terms).

byefruiton Elasticsearch does not belong to Elastic

I'm genuinely interested in examples of great projects from those funding sources.

When I think about OSS databases for example I can't think of any widely used ones that don't have at least one commercial entity that provides maintenance and support which funds development.

byefruiton Elasticsearch does not belong to Elastic

Does not being able to call itself Elasticsearch make a fork inoperable? I don't think wanting to protect a brand they've been investing in for years is unreasonable.

Don't get me wrong. It would be great if Elastic, Redis and co didn't have to do this (change license) but AWS does pose an existential threat to their business models long term. I'd rather a standardised, so there's less fear around interpretations, more restrictive license than the businesses and _future_ ones becoming unviable.

byefruiton Elasticsearch does not belong to Elastic

Is it? Do you really feel AWS' model is healthy for free software in the long term if business models around the creation and maintenance of open source projects disappear?

I would say Google's approach of partnering with maintainers instead of competing with them is healthier, though only just.

byefruiton Elasticsearch does not belong to Elastic

The post feels very self-entitled and needlessly inflammatory.

Elastic's changing their license. No contributions are being stripped of their open source license. You always have the last open source licensed version to go back to and fork if you like.

What this author is upset about is that Elastic are no longer providing updates under a license the author feels they are entitled to.

Maybe the author should fork the project and maintain it under an open license themselves?

byefruiton Surprising new feature in AMD Ryzen 3000

If you are relying on that behaviour then your code is already broken on most modern x86 processors due to their memory ordering implementations.

The store to the address at rsi will actually go in to the processor's store buffer and the subsequent add is allowed to retrieve the value from the store buffer _before_ it's drained to main memory.

Under TSO, AMD's memory mirroring implementation here is totally fine.

It could potentially change observable behaviour if you could turn it on or off but this is irrelevant given you wil see changes in observable behaviour for that same snippet of code depending on which microarchitecture of AMD or Intel processor you choose to run it on.

It's worth having a read on the details for different memory models and how they relate to implements. This seems like a good guide: https://www.cs.utexas.edu/~bornholt/post/memory-models.html

byefruiton Facebook apologizes to users for Apple’s monstrous efforts to protect privacy

I think you do get this argument and we're actually saying pretty much the same thing.

I'm saying people shouldn't look at this decision as "Apple has our interests in mind" but rather "in this particular case Apple's self interested is aligned with our own because of their business model".

byefruiton Facebook apologizes to users for Apple’s monstrous efforts to protect privacy

This isn't necessarily a principled campaign from Apple.

Apple no longer has an ad network which means it can't take it's cut from ad revenue. By reducing the effectiveness of ads on the platform it pushes more people to in app purchases which it receives a heft chunk of.

byefruiton Simdjson: Parsing gigabytes of JSON per second

This is close but incorrect.

Most common Lempel-Ziv compressors (gzip/brotli/zstd) use some form of entropy encoding (Huffman/Arithmetic/ANS) but that is in addition to lz algorithm (usually lz77) itself.

You can have a Lempel-Ziv compressor that doesn't use entropy encoding, look at something like lz4.

lz77 essentially works by copying parts of the recent window of decompressed data. e.g if a phrase of length 10 was used 100 characters ago we could encode 100,10 instead of the phrase itself. Most compressors use entropy encoding on those offset and length streams.

byefruiton Starting a Business Around GPT-3 Is a Bad Idea

Case in point: your business currently returns "Something went wrong: 'GPT3 API is currently overloaded, please try again momentarily.'"..

Do you have control over that?

byefruiton How to Run a Live Coding Stream on Twitch Using OBS

Love the bokeh on his video feed. Hard to justify a $900 camera for that alone though, anyone using anything else they'd recommend?

byefruiton Nvidia is now worth more than Intel

What makes you think it's not the right architecture for servers?

If anything AWS' Graviton2s are competitive on multithreaded workloads and it looks like the Neoverse cores are only going to improve there.