^ I think the biggest danger in rolling out AI across the enterprise will be the cheap managers.
They will lilkely converge towards the cheapest "solution" and that will reinforce the errors in the models. All AI designed speakers will become Bose.
Someday, when they're all using the Crowdstrike AI, everything will stop.
I recommend you keep a high quality AC generator at home... with a nice analog rig and plenty of cold beer for when that time comes.
They will lilkely converge towards the cheapest "solution" and that will reinforce the errors in the models. All AI designed speakers will become Bose.
Someday, when they're all using the Crowdstrike AI, everything will stop.
I recommend you keep a high quality AC generator at home... with a nice analog rig and plenty of cold beer for when that time comes.
Hi, I have a box, aprox. 70 liters. I have a driver, a Thomann 12-280/8-A cheapo. Now I need a speaker for passive usage as a stage monitor for drums (fullrange, ported) and/or a speaker for a bass amplifier (ported). Without corrections for bafflestep I will try to model it (and build it) using ChatGPT.
Stay tuned.
Stay tuned.
Some back up information as a sanity check maybe?
See here and maybe use as a point of comparison after ChatGPT has done its work.
https://www.hifi-selbstbau.de/index.php/hsb-datenblaetter/tieft/the-box-speaker-12-280-8-a?showall=1
See here and maybe use as a point of comparison after ChatGPT has done its work.
https://www.hifi-selbstbau.de/index.php/hsb-datenblaetter/tieft/the-box-speaker-12-280-8-a?showall=1
I detest links to sites that want to put cookies in my browser or who complain about my ad blocker.
Pffft...
Pffft...
Yes, I'm using their corrected T/S data, even if I'm suspecting that the tolerances are very broad in such a cheap series.
Why.
Ported speaker alignments are not too difficult.
Calculate the Box Volume, Calculate the Port.
Getting the box to look nice is most the challenge.
Something that is learned over time with hands on woodworking skills. And just collecting tools to make it easier.
Download WinIsd or Virtuix Cad and calculate a standard alignment.
The speaker suspension is either this stiff or that loose and the magnet is only so big.
It just needs the right volume to support that. Any speaker has a Fs ( free air resonant frequency). in the data sheet.
The port will tune a few Hz below or above that by 3 to 7 Hz depending if 3rd or 4th order.
Pretty simply. BB4 the most simple. Whatever Fs is, that is what the port is.
QB3 tune slightly above Fs C4 will tune slightly below Fs not more than 3 to 7 Hz.
Box size is made big enough to make speaker suspension happy.
Most people want live boxes to be light, so they use lighter plywood now.
The trick is to add bracing so it dont vibrate.
Dont make the box a perfect square. Slightly Rectangle.
To go over Loud drum kit usually you want 2x speaker or more for Bass Guitar 2x12 or 2x15, 4x10 or 4x12 the standard for years
Within reason live sound drivers will be around 1 to 1.5 cubic feet for 10" 1.8 to 3 cubic feet for 12" or 3 to 4 cubic feet 15"
easier to calculate according to specs. If the magnet is extremely powerful, it can go in smaller box.
You want small box then Qts should be below .4 anything .5 or higher is weaker magnet and will need a bigger box.
The only mistake is wanting to small of box and using weak driver. Then it is bloomy and boomy = Suspension not happy
Or tuning too high. Tune near or at Fs. Is that simple
Last edited:
The use of generative AI has been linked to decline in cognition, memory and executive function. If you willingly relinquish your mental abilities, you will lose them. The fact that millions of people seem eager to give up their intellects, to be freed from the burden of ever having to think or create, is a nightmare out of our darkest science fictions.
https://time.com/7295195/ai-chatgpt-google-learning-school/
https://time.com/7295195/ai-chatgpt-google-learning-school/
@gigantic That last sentence is an excellent expression describing the near future. I urge everyone to watch the 2006 movie Idiocracy.
https://en.wikipedia.org/wiki/Idiocracy
https://en.wikipedia.org/wiki/Idiocracy
The thread is not about the relative dumbing down of the developed world which happened well before this current "AI" bubble but about whether "AI" currently has a useful contribution to make to DIY speaker design. The answer seemed pretty much like no a few years ago whereas today it is seems more like a qualified maybe. The qualifications concern the amount of learning required to use it effectively, if you do commit to learn then you will likely find one or two of the components needed to be effective are not free and, of course, the only reason so much is free now is to get users dependent so that the billions invested can be recouped along with large profits from dependent users in the future.