Pictures for the weekend

[Wrath Of Khan]

Now we just need to see the a prototype with the finished case design and everything.

Can't be long now?

 

[DAP]

Since people seem to be interested in USB type-C, here are some links:

http://www.those.ch/designtechnik/wp-content/uploads/2014/08/USB-Type-C-Specification-Release-1.0.pdf

http://www.usb.org/developers/powerdelivery/PD_1.0_Introduction.pdf

http://www.usb.org/developers/docs/

The type C document is interesting. The cables have some intelligence, and in order to achieve 100 Watts (20V @ 5A) both the source and the cable must tell the load that they can handle it.

How the type C connector works as an audio jack is at the end of the type-C document.

 

[matzesu]

Well i ditnt have a portable SATA Drive so i wouldnd need the SATA Port on Pyra


Im happy whit the Option for a future USB 3..


Also its great that whe have that many USB Port whyle the new Mac Boock only have one..


Why ditnt whe just use the Volume Wheel from the Pandora?? It also works great and i cant image it wouldnt work..


Nice Pictures, and im happy the Rumble Motor is still there..

 

[Wrath Of Khan]

Get ready to RUMBLE...

 

[Glyph Reader]

I'm glad that the rumble is there if only for the possibilities it gives others, for I do not personally have an immediate use for it myself. 

-Glyph Reader

 

[levi]

The most obvious use for rumble is the emulators for systems that had rumble, surely?

Regarding the USB->SATA adaptor, will it have a way of signalling the board that it's connected?  I'm imaginging a future processor which has both an extra USB3 connection and a SATA connection, it'd be nice if we could still take advantage of both on that port.  So the adaptor would need to negotiate to select the SATA data lines, rather than the USB3 superspeed ones.  I guess it could just wire one of the USB2 data lines high or something like that, rather than needing to have any specific logic in it.

 

[ssokolow]

Why ditnt whe just use the Volume Wheel from the Pandora?? It also works great and i cant image it wouldnt work..

Probably because some users, myself included, report audio noise (What's the term for the "rubbing your finger over a microphone"-like sound caused by an electrical contact of inconsistent quality?) while the wheel is in the process of being moved.

 

[Silent-Hunter]

^I have this problem.

 

[thatgui]

As I never understood why there was eSATA in the first place, I'am more like this

https://www.youtube.com/embed/CxK_nA2iVXw?feature=oembedI experienced this myself (which was the reason why I opted for a digital solution in the beginning), but I thought a similar volume wheel will be used for the Pyra ?

@ED

seeing that the rumble motor will make it in the final product, have you decided upon the additional sensors Nikolaus included (Barometer n'stuff). The last info I remember about it was that you had not decided wether they will be included or not.

 

[EvilDragon]

^I have this problem.

Just spray a bit of contact cleaner into the volume wheel, rotate it a bit and that's fixed.

 

[Zink]

^I have this problem.

Just spray a bit of contact cleaner into the volume wheel, rotate it a bit and that's fixed.

I confirm. I had this problem twice already (approx. annually). The first time I used IPA cleaner, the second - anticorrosive spray. They both worked fine, although it seems the latter is better.

 

[_wb_]

Nice to have a future-proof USB3 port, even though the currently only CPU board can't handle it. It will lead to confusion though, with people expecting USB3 speeds and only getting USB2.

Is there a way to include a USB3 hub on the CPU board so both the micro-USB3 and the full-size USB3 port can actually achieve USB3 speeds? (though not both at the same time, since there would of course still be only one connection to the OMAP5)

That would be the most awesome solution, but I guess it is not possible to do something like that...

 

[slaeshjag]

No, since the USB interface can't be both an OTG device and a host device at the same time

 

[Slacker]

It will lead to confusion though, with people expecting USB3 speeds and only getting USB2.

As I said previously - make sure the port is of a different color than the characteristic USB3 shade of blue, and no confusion will happen. Problem solved.

Is there a way to include a USB3 hub on the CPU board so both the micro-USB3 and the full-size USB3 port can actually achieve USB3 speeds? (though not both at the same time, since there would of course still be only one connection to the OMAP5)

Abso-effin-lutely unpossible. The OTG port needs to run directly from the USB controller. If you put a hub between the controller and the port, the OTG functionality is lost.

 

[hns]

It will lead to confusion though, with people expecting USB3 speeds and only getting USB2.

As I said previously - make sure the port is of a different color than the characteristic USB3 shade of blue, and no confusion will happen. Problem solved.

Unfortunately we can't choose/change the color of the socket (they come in blue from the factory and to order a different color we need to buy quite hug he quantities).
Next, this limitation is not by the connector or the main board, but the OMAP5 CPU board. The OMAP5 was simply designed when USB3 wasn't that common but eSATA so they decided to have only one USB3 (OTG) port instead of two and add eSATA.

Now, our CPU board is exchangeable. What would happen if we can offer a future CPU with full USB3? Someone would have to change the color of the socket as well...

IMHO the problem is smaller than you think. First of all, there will be the USB3-socket to eSATA socket adaptor in the list of accessories.

Secondly, what happens if you connect a USB3 capable device? It will simply run at USB2 speed which is still ca. 48 MByte/s (read or write).

If that is really not enough, connect it to the micro-USB OTG port.

Therefore I think we have to live with this small imperfection until a real USB3 CPU board can be built.

And no, there is no space for an USB hub and you are right we would loose OTG capabilities. It would also draw additional power.

 

[Binky]

If a future SOC can do both eSATA and USB3.0, will it be possible to switch the sockets behavior in software (i.e chose between USB3.0 or USB2.0 + eSATA-via-adaptor)

 

[hns]

If a future SOC can do both eSATA and USB3.0, will it be possible to switch the sockets behavior in software (i.e chose between USB3.0 or USB2.0 + eSATA-via-adaptor)

That is unlikely. They must have a combined and switchable PHY on SoC (that could be possible since eSATA and USB3 are quite similar!). Or we would have to add external RF switches capable of carefully switching 6 GHz signals.The question is how long eSATA will be of interest at all. I am under the impression that it becomes already difficult to find devices with external eSATA. And that are hard disks and hard disk adapters only. All other devices come with USB3, even many HDD cases.

And, I think there are (real) USB3 to eSATA adapters available. Real means that they do signal and protocol conversion. Not just wiring.

 

[WizardStan]

The question is how long eSATA will be of interest at all.

It's irrelevant, there's no practical (speed) benefit to using an eSATA harddrive vs a USB3.0 one. Once we've got the capability of dedicated USB3.0 host you can ditch the eSATA port.

 

[levi]

If a future SOC can do both eSATA and USB3.0, will it be possible to switch the sockets behavior in software (i.e chose between USB3.0 or USB2.0 + eSATA-via-adaptor)

That's why I suggested a signal the USB->eSATA adaptor can make to the board to signify it's to connect the SATA lines to the extra pins, not the USB3 superspeed pins.

I'm not sure if anything needs to be done now - it depends if the switching needs to be done on the peripheral (main) board or the CPU board I think.  I don't really understand how the peripheral board -> CPU board connectors interface the physical external pins to the CPU pins - if they are just direct connections, with each pin of the USB port both hi speed and super speed going directly to pins on the CPU connector, then I guess whatever sensor circuit can be designed into a future CPU board that supports both >1 USB3 port and eSATA, should such a board ever get made.

But it would need the adaptor being made to release alongside the OMAP5 Pyra to apply this signal, so that a future CPU board can react to it.

 

[DAP]

What happens when we get a SOC that integrates the USB super speed pin switching required by the type-C connector? Does the SOC connector pinout support 8 wires for the 4 differential pairs associated with the USB type-C connector?