Recent Posts

1

AI / Re: AI destroying economic system

« Last post by tszaboo on Today at 03:01:58 pm »

Another option is to fire the employees that make these decisions, and write efficient software.

For what platform?

What platform are you writing it for?

I think Microsoft had a very good idea where people are going to run their Windows built-in weather app.

They are getting closer and closer to convincingly real world looking, in real time renders.  If you were to half the RAM budget you would see a significant drop in texture quality and or frame lag/speed/rate. 

Graphics has been degrading for years now. They just use AI upscaling and other fake methods to make their shitty game engine working on consoles. New games come out at 900p upscaled to 1080 because they can't hit 30 FPS mark on a console that came out years ago, and well understood.

EDIT:  For reference.  That game download size, with the addons I have is 450Gb!

You are saying it like it's a good thing.

2

Repair / Re: Agilent 6612C: Strange noise, what could that be?

« Last post by hp3310a on Today at 03:01:55 pm »

Well, I can confirm that the connector wasn't the culprit. I desoldered it and tested again, basically same result. Noise is a little lower but that's due to lights being switched off. Noise gets even lower when using the probe with a ground spring. But the comb-like noise that I'm wondering about is still there.

3

General Technical Chat / Re: Blood-drawing robot

« Last post by nctnico on Today at 03:01:09 pm »

Or it can be used in prisons on death row inmates, to more successfully identify suitable veins for lethal injection.

Just put them in a mostly sealed enclosure with a constant feed of nitrogen. The CO2 gets flushed away, which would prevent the feeling of suffocation. All that happens is that they drop dead fairly peacefully as they're no longer getting oxygen, just like what happened in accidents caused by oxygen displacement or gradual depressurization.

Or accept that the price of living in society is that bad people go to prison for the rest of their lives, and the state shouldn't be executing people unless it is the only viable option - an immediate decision to prevent the further loss of life e.g. an active shooter.

The death penalty seems to come from bloodlust and I can see why people might feel it's the only acceptable outcome but ultimately it costs more than keeping someone in prison for the rest of their lives (look at how long cases drag on - inmates face about 20 years of hearings before any final decision is reached, that's a lot of expensive legal arguments) and it never brings the victims back.  And it clearly hasn't reduced the rate of murders given states that have no death penalty see vastly fewer murders, at the very least it could be argued there is effectively no correlation.

I agree. Killing somebody for a crime is never the answer. However, lifelong imprisonment is considered inhumane as well. It is basically locking somebody up in a hole waiting for the person to die.

Back to the blood drawing robot: it could be an interesting solution. Drawing blood is not an easy skill to master. If the robot can do equal or better compared to a skilled nurse, then it can add value for sure.

4

Microcontrollers / Re: How to get a random seed for a random number generator?

« Last post by ledtester on Today at 03:01:08 pm »

The good old days, when Nintendo secured QA-approved NES cartridges with a custom 4-bit chip...

I was curious so I researched this and it's quite a fascinating story!

Some of the main points:

- In the late 80's Nintendo develops the 10NES lockout system and patents it:

The Nintendo 10NES (CIC) lockout system is covered by U.S. Patent 4,799,635, titled:

"System for Determining Authenticity of an External Memory Used in an Information Processing Apparatus."

https://patents.google.com/patent/US4799635

- Two identical chips, one on the NES and one on the cartridge, run in lockstep producing the exact same sequence. If the sequences differ the NES chip resets the CPU.

- Tengen, a subsidiary of Atari, using subterfuge gets the source code for the 10NES from the US Copyright Office and this leads to the development of a functional substitute they call the "Rabbit". Nintendo sues and in 1992 wins an injunction against Tengen.

- The Action 52 cartridge used voltage fault injection to disable the 10NES chip. A charge pump was used to create -5V which was presented to the 10NES data lines. Nintendo later adds resistors and clamping diodes on those lines.

- Another approach to running unlicensed cartridges was to piggyback a genuine cartridge via a dongle which passed through the 10NES signals from the genuine cartridge. Yet another was to cutting the connection from the 10NES chip in the NES and the 6502's reset line. Later Nintendo games start to check for this modification.

- In 2006 the Rabbit chip is decapped and although the CPU itself isn't reverse engineered it is discovered that the Rabbit has a debug port which allows introspection of what each instruction is doing, and this leads to the full revelation of the 10NES algorithm.

There are tons of articles/videos on this saga. For those who like to read check out pages 6-9 here:

https://archive.nes.science/nintendoage-forums/static.nintendoagemedia.com/ezine/Issue_07-2008.pdf

5

General Technical Chat / Re: Flying Tesla Roadster

« Last post by Randy222 on Today at 02:59:18 pm »

There's no argument to be had about ICE vs E efficiency.

Today, the argument is more about not contributing to greenhouse effect.

Dirty petro ICE is at best 41% efficient, but very cheap energy source. HICE may be even less efficient but because it produces way less atmospheric pollutants we don't care so much about it being less efficient.

FCEV is the real solution to the problem. H is the obvious energy source. H can be made 24/7 by breeder reactors, and, via daytime solar power.

There's certainly argument to be had for the all-batt EV if (if) the batt component can be made/replaced/recycled/disposed in an eco non-toxic manner. Which today the Lithium stuff is very far from that.

The rest of the problems around H or electric grid capacity, are just money issues. I have not seen Elon take some of his $B to upgrade a power grid anywhere. If he's selling EVs then why not help fix the problem that EVs need fixing? He could do it in partnership with GM/Ford/VW/Hyundai, as example. Yet for some reason there seems to be no obvious desire to tackle the grid problem, by anyone.

6

Test Equipment / Re: HP856xE hackers' corner

« Last post by jjoonathan on Today at 02:52:40 pm »

I open sourced python notebooks and made a youtube video walking through the tracking and flatness calibration of 856xE analyzers a couple of years ago. I am disappointed that claude did not find it, although perhaps it got lost down the garden path of reading the official procedures that, of course, depend on the TAM.

Github: https://github.com/jjoonathan/8560E-Calibration

Youtube:

7

Test Equipment / Re: NovaOS Firmware for Rigol DHO800/900 Oscilloscopes

« Last post by bradyok on Today at 02:31:20 pm »

And I'm flying to meet with Rigol to discuss everything, just because I am selling their scopes doesn't mean we have agreed on all things related to the software, there is a lot to discuss!

While you're there, you might also want to ask Rigol why your company isn't yet listed on their official North American distributor's list. 

https://www.rigolna.com/company/distributor/

Your website also doesn't mention being an authorized reseller, so some people might be nervous about purchasing Rigol scopes from your company.  Generally speaking, warranty service is only available if you buy instruments from an authorized, approved reseller, so might be good to get them to fix this for you

Nothing is available for purchase yet, I haven't received my first shipment yet, my website only has the ability to request a quote. The quantities provided on the website are the Rigol USA office stock. These things take time, I filled out paperwork, wired money, now I'm waiting patiently. And yes, I have not claimed to be an official distributor until their website reflects this.

8

AI / Re: Alibaba's Qwen phones home

« Last post by PlainName on Today at 02:31:12 pm »

has anybody managed to confirm the claims in the articles which to me smell like FUD to promote ...

That's a fair question and, of course, no-one here has. We have to take a certain amount on trust so rely on our sources being probably straight up.

9

Repair / Re: The Yokogawa DL1740E has a gray screen and won't turn on.

« Last post by TinLethax on Today at 02:30:37 pm »

I didn't install the entire front board, powered it on for testing, and it's still a black screen. The A6236LG is insanely hot, and there are four ICs on the back. Seems like they run on 3.3v, and all the heating ICs are taken off. There are pictures later. (Attachment Link)

Well, my scope also suffered the A6236LG (Trigger chip) failure. I managed to source it from Alibaba. But you can find it on Taobao also.
Mine shorts from -5V to 3.3. You can refer to my personal blog posted by GGMM.

10

Projects, Designs, and Technical Stuff / Re: Redesigning a ±30A Bipolar Current Source – Looking for Advice on Circuit Design

« Last post by mawyatt on Today at 02:27:46 pm »

Take a look at the modified Howland Current source like we are using in our DIY Curve Tracer. This should be able to scale to your current ranges. Note how we use the class B bias by allowing the core amp to deliver the lower current ranges, while the Darlington output transistors to supply the supplemental higher current levels. This is intended for low frequency use tho, as the Curve Tracer operates at a low frequency range.

https://www.eevblog.com/forum/projects/gp-curve-tracer-for-cro-and-dso-use/

Best