tchiwam
Posts: 43
Joined: Mon Nov 24, 2014 4:01 pm

Re: What do people want to do that the rpi currently cannot

Fri Dec 18, 2015 2:20 pm

I can't have a full sky camera, 1/4" is simply too small for the 1.04mm C mount lenses to give the full 185deg. Also preventing spectroscopy. A Mono sensor 1/3", low resolution would be fun for showing light physic. But the work around is to project on a 2nd plane.

A mic input, Hard to show analogue signals without having to add an AD, work around is to buy a cheap USB sound card.

A better wide input power supply, allows to lower the Amps, power a display/RF/... and the PI from a single source on the power strip, becomes important when you have 15 kids. And saves money too by not trying to find good quality and safe >2A-5V. We'd like to see the kids go home and have fun with it and not burn down the house(Aka wierd smelling "2A" 5V power from far east).

The display is the biggest cost of any combination of kits right now, that's a pretty hard limit when you multiply the cost x 15 or more...

I've always found a work around, so far...

W. H. Heydt
Posts: 13616
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: What do people want to do that the rpi currently cannot

Fri Dec 18, 2015 3:34 pm

tchiwam wrote:I can't have a full sky camera, 1/4" is simply too small for the 1.04mm C mount lenses to give the full 185deg. Also preventing spectroscopy. A Mono sensor 1/3", low resolution would be fun for showing light physic. But the work around is to project on a 2nd plane.
So you want a fisheye lens. That sort of thing would be the point of having a lens mount, and if C-mount is too big, D-mount (which was used for standard 8mm cine cameras) might be the better choice, though you're less likely to find really specialized lens in that mount. FYI...a "normal" D-mount for cine work is 12.5mm focal length. Given the focal length of your C-mount fisheye, would suggest that a D-mount one would be about 0.5mm.
The display is the biggest cost of any combination of kits right now, that's a pretty hard limit when you multiply the cost x 15 or more...

I've always found a work around, so far...
I have a use where I use a bunch of monitors (though not 15, so far, thankfully) and I, too, have found ways to deal with this issue.

My hope here is that new tech will emerge (such as "ink jet" printed OLED) that will permit making reasonable monitors (say, 12" to 13") much cheaper ($10-$15?) and that the RPF will make the arrangements, with a bit of a royalty coming back to sustain their efforts.

KeithSloan
Posts: 321
Joined: Tue Dec 27, 2011 9:09 pm

Re: What do people want to do that the rpi currently cannot

Sat Dec 19, 2015 5:40 pm

I would like the pi to be able to access Skype & Dropbox. Both are available on x86 Linux but not open source so the owners of the software would have to be persuaded to compile for Arm.

I would have thought that recompile of Dropbox would be relativity straight forward.

Skype might be more effort to support the Pi's camera and a microphone. Skype is now owned by Microsoft and at least they know the pi exists.

Is there anyway we can barrage these companies with support requests

tchiwam
Posts: 43
Joined: Mon Nov 24, 2014 4:01 pm

Re: What do people want to do that the rpi currently cannot

Sat Dec 19, 2015 6:01 pm

W. H. Heydt wrote:
tchiwam wrote:I can't have a full sky camera, 1/4" is simply too small for the 1.04mm C mount lenses to give the full 185deg. Also preventing spectroscopy. A Mono sensor 1/3", low resolution would be fun for showing light physic. But the work around is to project on a 2nd plane.
So you want a fisheye lens. That sort of thing would be the point of having a lens mount, and if C-mount is too big, D-mount (which was used for standard 8mm cine cameras) might be the better choice, though you're less likely to find really specialized lens in that mount. FYI...a "normal" D-mount for cine work is 12.5mm focal length. Given the focal length of your C-mount fisheye, would suggest that a D-mount one would be about 0.5mm.

My issue is I do need the sub f1.4 for Northern light photography.

davenull
Posts: 1159
Joined: Thu Oct 22, 2015 7:22 am
Location: a small planet close to Betelgeuze

Re: What do people want to do that the rpi currently cannot

Sat Dec 19, 2015 6:17 pm

What do people want to do that the rpi currently cannot do?
I would like the RPi to be able to access GPIOs as quickly as digital pins by Arduino, by direct hardware access and also providing ARM timer interrupts, not by Linux user space.
Additionally, all GPIO pins should be capable of hardware pwm.

edit:
Not so urgently important, but nice would be nevertheless, as just 16 digital pins is extremely poor:
32 digital and 8 ADC pins and 1 DAC pin would be wishful, too.
Also a 2nd i2c and a 2nd UART port would be great.
Last edited by davenull on Sat Dec 19, 2015 6:23 pm, edited 2 times in total.
#define S sqrt(t+2*i*i)<2
#define F(a,b) for(a=0;a<b;++a)
float x,y,r,i,s,j,t,n;int main(){F(y,64){F(x,99){r=i=t=0;s=x/33-2;j=y/32-1;F(n,50&S){t=r*r-i*i;i=2*r*i+j;r=t+s;}if(S){PointOut(x,y);}}}for(;;);}

W. H. Heydt
Posts: 13616
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: What do people want to do that the rpi currently cannot

Sat Dec 19, 2015 6:22 pm

tchiwam wrote: My issue is I do need the sub f1.4 for Northern light photography.
That adds complications. It's generally hard to find really fast lenses when you start moving away from "normal" ones. That said, one used to be able to find 25mm (normal) lenses for 16mm cine work (C-mount) at F0.95. The fastest lenses for 35mm were F1.0. Pretty much everybody made an F1.2 and/or F1.4 50mm for 35mm cameras.

On the good side...since cine film was generally slow, and shutter speeds tightly constrained, on average, cine lenses are a lot faster than the ones for still photography. There are (or were) lots of lenses in the F2 to F2.8 range. One of the ones I have is a 152mm (6") F2.7 C-mount, for instance. Just try to find the equivalent to that for 35mm work without costing a fortune (it would be about 600mm for the same coverage--my 600mm cat is F8, for instance, and probably about T10).

tchiwam
Posts: 43
Joined: Mon Nov 24, 2014 4:01 pm

Re: What do people want to do that the rpi currently cannot

Mon Dec 21, 2015 5:37 pm

W. H. Heydt wrote:
tchiwam wrote: My issue is I do need the sub f1.4 for Northern light photography.
That adds complications. It's generally hard to find really fast lenses when you start moving away from "normal" ones. That said, one used to be able to find 25mm (normal) lenses for 16mm cine work (C-mount) at F0.95. The fastest lenses for 35mm were F1.0. Pretty much everybody made an F1.2 and/or F1.4 50mm for 35mm cameras.

On the good side...since cine film was generally slow, and shutter speeds tightly constrained, on average, cine lenses are a lot faster than the ones for still photography. There are (or were) lots of lenses in the F2 to F2.8 range. One of the ones I have is a 152mm (6") F2.7 C-mount, for instance. Just try to find the equivalent to that for 35mm work without costing a fortune (it would be about 600mm for the same coverage--my 600mm cat is F8, for instance, and probably about T10).
The problem with northern lights, we need wide and fast :)

davenull
Posts: 1159
Joined: Thu Oct 22, 2015 7:22 am
Location: a small planet close to Betelgeuze

Re: What do people want to do that the rpi currently cannot

Mon Dec 21, 2015 6:44 pm

is "Northern light photography" a Raspi wish issue?
#define S sqrt(t+2*i*i)<2
#define F(a,b) for(a=0;a<b;++a)
float x,y,r,i,s,j,t,n;int main(){F(y,64){F(x,99){r=i=t=0;s=x/33-2;j=y/32-1;F(n,50&S){t=r*r-i*i;i=2*r*i+j;r=t+s;}if(S){PointOut(x,y);}}}for(;;);}

yoshiking24
Posts: 1
Joined: Tue Jan 19, 2016 4:28 am

Re: What do people want to do that the rpi currently cannot

Tue Jan 19, 2016 4:35 am

I think there should be native support for Android and Windows (with full GUI), sideways GPI/O pins, built-in WiFi, and the USB ports spread along the side, and some mini-Ethernet or something, so it is slimmer. So basically, slimmer and faster.

davenull
Posts: 1159
Joined: Thu Oct 22, 2015 7:22 am
Location: a small planet close to Betelgeuze

Re: What do people want to do that the rpi currently cannot

Tue Jan 19, 2016 3:23 pm

more Pins would be fine (digital I/O, pwm, ADC, DAC, UART) -
if there is a need for keeping the 40 pin header, one could drop some 3.3V, 5V and especially GND pins and re-use them for signals instead.
Already existing I/O pins could be enhanced to support pwm or communication, too.
Otherwise, the 40 pin header could be enlarged or extended to (e.g.) 50+ ...
#define S sqrt(t+2*i*i)<2
#define F(a,b) for(a=0;a<b;++a)
float x,y,r,i,s,j,t,n;int main(){F(y,64){F(x,99){r=i=t=0;s=x/33-2;j=y/32-1;F(n,50&S){t=r*r-i*i;i=2*r*i+j;r=t+s;}if(S){PointOut(x,y);}}}for(;;);}

W. H. Heydt
Posts: 13616
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: What do people want to do that the rpi currently cannot

Tue Jan 19, 2016 5:29 pm

yoshiking24 wrote:I think there should be native support for Android and Windows (with full GUI), sideways GPI/O pins, built-in WiFi, and the USB ports spread along the side, and some mini-Ethernet or something, so it is slimmer. So basically, slimmer and faster.
For MS Windows, the Pi would have to shift to an x86 processor. That would entail throwing out all the software work that has been done over the last 4+ years and starting over. It would also mean far higher power requirements and much greater cost. Really not a practical or realistic move for the RPF.

And besides...the Pi as it stands supports windows in that the default distro includes a GUI desktop.

toddmaxwell
Posts: 1
Joined: Tue Jan 19, 2016 5:32 pm
Location: Lakeland, Florida
Contact: Website

Re: What do people want to do that the rpi currently cannot

Tue Jan 19, 2016 5:37 pm

Raspberry software works quite well on our automated gate. Never even heard of it until a few months ago

W. H. Heydt
Posts: 13616
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: What do people want to do that the rpi currently cannot

Tue Jan 19, 2016 11:08 pm

This is going to look a little silly (and, in a way, it's a take off of the Major General's song in the G&S operetta _The Pirates of Penzance_), but here goes...

What I want is:
the hardware design quality, software support, and USB port count of the Pi B+/Pi2B
the processor word size of the PINE64+
the clock speed of the Odroid-C1/C1+
the memory size of the Roseapple Pi or the top version of the PINE64+
the mass storage interface of the Cubieboard or Banana Pi
the built in RTC of the PINE64+ or Odroid-C1/C1+
the independent Ethernet bus of pretty any of those except the Pi

You can see that the features I want all exist, just not all on a single board. There are a couple of places where a compromise would still be okay. Clock speed could be that of the PINE64+ or Roseapple Pi and the USB count could be that of any of the non-Pi boards (IIRC all of them have at least 2).

davenull
Posts: 1159
Joined: Thu Oct 22, 2015 7:22 am
Location: a small planet close to Betelgeuze

Re: What do people want to do that the rpi currently cannot

Wed Jan 20, 2016 8:51 am

CMIIW, but it's just a poll about "What do people want to do that the Raspi2 currently cannot do", it's not about "what the hell does all exist in the entire universe which has to be indispensibly integrated to the next Raspi3"
IMO it's never wrong to ask users what would be wishful to have for the future (if not even for the future which one would call "near" though, perhaps)
Finally technics proceed and times are a-changing... ;)
#define S sqrt(t+2*i*i)<2
#define F(a,b) for(a=0;a<b;++a)
float x,y,r,i,s,j,t,n;int main(){F(y,64){F(x,99){r=i=t=0;s=x/33-2;j=y/32-1;F(n,50&S){t=r*r-i*i;i=2*r*i+j;r=t+s;}if(S){PointOut(x,y);}}}for(;;);}

Return to “General discussion”