foft

The question is what is reasonable. This is more how I work for code and I just tried for a document.

i) Prompt (Human input)

ii) Response (LLM)

iii) For (i=0;i!=n;++i) {

iv) Review/adjust (Human input)

v) Response (LLM(s) - different perspectives) }

vi) Rewrite parts that failed/I am not happy with (Human input).

vii) Optionally repeat from iii.

This tends to get me somewhere I want. For code at least. I made a document with it recently (fairly concise) and still worried I'd missed some AI slop in it. Though there is always human slop if I just made it by myself.

Cutting coffee to just the morning definitely helped me.

My problem now is that if I think of anything vaguely interesting in the evening then my brain refuses to stop thinking about it. That and restless legs.

Evening if my main time for hobby projects so having them lead to disrupted sleep is really annoying. Any tips?

Really awesome. Though I wish they'd do a dense 48B, 60B or 72B.

There seems to be quite a gap between the small ones and the enormous ones these days.

If you put your data in Snowflake then there is a built in AI (ok it’s Claude) that can access the databases. This sidesteps a lot of the issues in that the data is clearly already with Snowflake.

Having used codex a fair bit I find it really struggles with … almost anything. However using the equivalent chat gpt model is fantastic. I guess it’s a matter of focus and being provided with a smaller set of code to tackle.

Yes. If you haven’t yet read it Cory Doctorow’s new book Enshittification is well worth a read. I am still reading it but it certainly explains some of the bad practices by these major advertising/spying giants and the resulting market distortion. We need to up our game as technologists and hold our employers to account.

It was a strange decision to limit the drives. I can see they might want to accredit drives which would give a 'Synology Approved Experience', though outright only support their own was bizarre. I'm very pleased they are reversing this. Aside: Now we just need Apple to do the same and resume support for industry standard expandable memory and storage.

From my perspective it lined up exactly with when I was looking to upgrade. I decided to bite the bullet and go with Duplicati, storing to a European based S3 service. I decided against US cloud providers since the US is looking too politically unstable to put anything important there. It was easy to set up and so far is running well.

If you are interested in FPGA PCIe or PCI cards there are also a lot of Gidel boards available second hand. ProcSpark/ProcStar etc. The official software is proprietary and they often contain multiple FPGA devices so there is again an exercise working out the pinouts etc if you want to simply use them with Quartus. I got one with a massive Stratix IV on.

A Kintel UltraScale+ is quite a prize though, really nice write up.

As a big openscad fan I love the idea of designing circuits with code.

I do wonder though about designing circuits vs designing schematics. I see you have ‘wire down 100’ making it a more visual language than defining the nets. Be interesting to separate the schematic layout from the nets, so rule base schematic layout can then be applied.

I've had great results with Codex, though I found ChatGPT 5 was giving much better results than the existing model. So ended up using that directly instead. So very excited to have the model upgraded in Codex itself.

The main issues with Codex now seem to be the very poor stability (it seems to be down almost 50% of the time) and lack of custom containers. Hoping those get solved soon, particularly the stability.

I also wonder where the price will end up, it currently seems unsustainably cheap.

In Switzerland pretty much all sites have a setting for French, German, Italian and often English. It is very much a multi-language country.

It is great to be able to select individually per site. I often like to use the native French and just drop back to English if its technical language.

I do find that every site has the setting in a different place which is annoying, it would be great to be able to select it in a standard place on the browser.

The worst offenders are the single language per country sites. For example Ebay insists on only using German in Switzerland, which is rather frustrating since I only know English and French so far.

If anyone was interested in the Amiga but has not kept up with recent developments, I suggest looking up the Vampire V4. It is mentioned in the article but I thought I'd add a few more details.

It has a reimplementation of the ECS and AGA chipset. It includes custom extensions to the chipset to 'SAGA' which is an attempt at extending the registers to more modern standards.

It also has a reimplementation of the 680x0 CPU, which is using more modern design techniques. The developer used to work on Power.

Anyway putting it all together its a great system in the vein of the Amiga. Of course it is not as fast as a modern ASIC, being consumer low end FPGA based. Still it is great fun.

Relevant to the Amiga 600? Well there is a standalone version but there is also a version called 'Manticore' that fits into the Amiga 600.

Many people will say you can get similar performance with emulation. This is of course true though, as someone who studied microelectronics, I see the value in real hardware. Both in future potential for making an ASIC and for more precise sub-microsecond level timing.

There is an alternative semi-emulation approach. i.e. emulating the CPU with a raspberry pi and using the rest of the original hardware. This is known as PiStorm and connects the GPIOs from PI onto the 68K to replace the original CPU.

Ouch, if that is the case this is only going to boost the non US suppliers. Has anyone tried lcsc.com? Digikey and mouser are great but if all semiconductors go up 50% that is a problem.