So, what's the current status of everything?


This is traditionally about the worst/busiest times for (small) businesses. I imagine he's working his socks off, but I trust he will indeed do everything possible to prioritize getting those first units out asap.

I just hope Grench (and co) will be able to update us as and when they get their prototypes :)
 
Wasnt it so, that whe waiting for the Cases to be ready in January, and then Massproduktion and Shipping??,
So EvilDragon cant now make anything to hurry up the Pyra at this point??
 
But the Protoypes are also important: , Prototyphe Preorderers can help whit the development of the Software, as they have the Hardware to test, so the Software in the MP Units are more improved,

And a Protoyphe Preeorderer can make lots of Pictures and Videos, and spread the word to others, so more Pyras can be selled ..
 
I just hope Grench (and co) will be able to update us as and when they get their prototypes :)

No updates to be had. The last status I saw was publicly posted. ED said that there were minor bits on the prototype mainboards that had to be altered. The work had been done to the units that went out to devs but had not been done to any of the boards with the US 4G modems on them - and apparently all of us prototype pre-pre orders are US 4G requests. I understand that the de-soldering/movement/replacement/soldering of components on this scale is excruciating detailed time consuming work. So, I'm trying to just sit back and be patient.
 
it would be best to assume nothing is getting done while the holidays are in full buy-everything mode (ED is very busy with the shop). if something does get done, it can be a pleasant surprise.
 
it would be best to assume nothing is getting done while the holidays are in full buy-everything mode (ED is very busy with the shop). if something does get done, it can be a pleasant surprise.
Why the pessimistic view? Is there any reason to think that nothing is being done on the Pyra?
 
Indeed, a view of the kernel mailing list at least shows there's more than ED working on it. Although last I heard, hns was trying to read up on the audio subsystem to start to debug it, although he'd coming from basically a very low start, so I'm not expecting any news from that front just yet.
 
Well, it's not strictly not working, it does create devices for the speakers and the headphone port and they work (as long as you don't overdrive them), but it also creates some extra fake devices that crash the kernel if you try to access them, which I'd imagine breaks pretty essential things like alsamixer and pulseaudio configuration screens.

Edit: And thus I can imagine benefits from sending out prototypes to testers and devlopers. At least it'll get more people noticing what's broken and making noise about it, even if none of them can help explicitly in fixing it.
 
Thos extra audio devices are not a problem though, as these exist as a result to try to load the firmware. If we use the mainline driver, these won't be there.

Just a quick info about the driver:
The driver is working and we have audio both on the headset and speakers.
We need to fix some parameters in the driver (dc offset and audio gain) so that it doesn't overdrive.

The only real problem with the driver is, that it doesn't support the hardware acceleration right now.
The last known kernel where this still worked is v3.14. Since then, a lot of api changes have been done to the kernel but the ABE / AESS part (this is the hardware acceleration support) has not been updated.
This needs to be done.

The audio chip only supports 96kHz 24 bit audio. So anything that has a different frequency (which is basically everything) needs to be samplerate converted.
With the current driver, this is happening in software (and thus pulseaudio used 30% CPU when playing with Drastic).
The ABE / AESS support would do this in hardware, so no CPU power used.

That's all :) Sound is working, just needs some CPU power.

Oh, and BTW: Yep, I didn't post because I was horribly busy with work. Basically, I didn't even find the time to play some games within the last week.
Preparing and shipping for christmas, organizing some Pyra stuff, finishing taxes, producing 2100 UltraCore MegaDrive carts, etc.

The parts I need to work on the prototypes have arrived just today. I've reserved the full weekend to test and fix any issues with the boards so I'll hopefully have these ready end of the weekend to prepare the prototype shipment.
Yep, we're close to christmas, but I try to finish them.

As long as I don't overdo it and fall asleep during work (happened 2 or 3 times within the last weeks), I should be good.

The good news is: I'm not dead yet :D
 
Just a quick info about the driver:
The driver is working and we have audio both on the headset and speakers.
We need to fix some parameters in the driver (dc offset and audio gain) so that it doesn't overdrive.

The only real problem with the driver is, that it doesn't support the hardware acceleration right now.
The last known kernel where this still worked is v3.14. Since then, a lot of api changes have been done to the kernel but the ABE / AESS part (this is the hardware acceleration support) has not been updated.
This needs to be done.

The audio chip only supports 96kHz 24 bit audio. So anything that has a different frequency (which is basically everything) needs to be samplerate converted.
With the current driver, this is happening in software (and thus pulseaudio used 30% CPU when playing with Drastic).
The ABE / AESS support would do this in hardware, so no CPU power used.

That's all :) Sound is working, just needs some CPU power.
Sounds good. *badumtss*
 
The only real problem with the driver is, that it doesn't support the hardware acceleration right now.
The last known kernel where this still worked is v3.14. Since then, a lot of api changes have been done to the kernel but the ABE / AESS part (this is the hardware acceleration support) has not been updated.
This needs to be done.

The audio chip only supports 96kHz 24 bit audio. So anything that has a different frequency (which is basically everything) needs to be samplerate converted.
With the current driver, this is happening in software (and thus pulseaudio used 30% CPU when playing with Drastic).
The ABE / AESS support would do this in hardware, so no CPU power used.

..some of this may be a bit over my head, or perhaps I'm miss-interpreting something, but does this mean the audio chip can currently only support nothing but 96kHz 24 bit audio because certain parts of the api aren't quite up-to-date or because the audio chip can't handle anything more/greater than 96kHz 24 bit audio without down-converting it? this sounds troubling...
 
Is it this chip ?
image.jpg


16 bit 44.1 kHz is CD quality audio ( I believe ), so 96 kHz 24 bit should be good enough for most music ( which is the best DVD-Video can handle I think ).
 
Last edited:
Yes, it should be 'enough' but it still takes significant CPU time to interpolate between 44.1 and 96kHz. 48 goes into 96kHz a lot easier (just repeat each sample twice), which I guess is why most audio these days is 48kHz while the 44.1kHz that was designed to adapt well to the human ear is less common. It may be good enough to repeat the sample twice most times but repeat each 10th sample three times, but that'd be out by just over 1/100, such that a 3 minute pop song would end up being 1.8 second too long, which might cause awkwardness in games that play their samples sequentially (and it would also alter the pitch of the samples, but I'm not sure what 1/100 sounds like as a pitch delta off the top of my head)
 
Back
Top