PDP2011 new release

[Sytse van Slooten]

Hi all,

I've just posted a new release of PDP2011 on https://pdp2011.sytse.net/ and there also you'll find a post describing the changes (there are quite a few). I'll cut and paste the list here for your convenience:

Base system

• The RH disk controller used to only work with RP06 images. It now also works with RP04/RP05, RM05, and RP07. Also, it supports the Setasi/Shelby RP12 RM06 layout of 64 sectors, 32 tracks that will work with RSX-11MPlus for disk images up to 2GByte (with thanks to Johnny Billquist for the info and help), and a synthetic 64 sector, 64 track drive that works with 2.11BSD. The controller still only supports one disk drive at a time though (and I don't think I'll fix that soon).
• Images for the 'artificial' drive types (RM06, RP2G) can be created by a modified version of simh that supports the new drive geometry. This has been tested up to a size of 2Gb (larger will not currently work). I haven't published the required (relatively minor) changes yet, but will try to do so soon.
• The serial link driver did not work very well if it would receive short stop bits from the other end - potentially leading to garbled characters. It should now be lots more stable and capable of higher speed. The issue was analyzed by Steve Falco (thanks!).

MINC

• MINC now supports up to 64 A/D channels; the A/D channels present on the board are the lowest numbered channels, and all higher numbered channels are available to insert values into MINC that are obtained from other sources like sensors, other types of A/D converter, computed data, etc - as long as they more or less fit in a 12-bit signed integer.
• All MNC module types can be configured to the maximum the software can handle: 1 MNCAD, 2 MNCKW, 4 MNCAA, 4 MNCDI and 4 MNCDO. That means 16 DA channels, and 64 digital input and output channels - plus some control signals.
• The logic in MNCDI for detecting state changes has been corrected; the previous version would only detect rising edge changes.
• A driver for the PMOD DA4 (Analog Devices AD5628-1) has been added, next to the PMOD DA2 that was already available.
• A driver for the PMOD NAV (Microchips LSM9DS1 and LPS25HB). The digital barometer in LPS25HB will likely feature in a future project for a weather station.
• A driver for the PMOD HYGRO (TI HDC1080). The temperature and humidity sensors will also be useful for a future weather station.
• A driver for the PMOD COLOR (AMS TCS3472). This sensor allows to determine the color temperature, and hence elegantly establish if the sun is shining or if there are clouds. 
• A driver for the ADXL345 accelerometer that is present on the DE0-Nano boards.

VT100/VT105

• The terminals now support double width and double height characters.
• Automatic XOFF/XON has been added, and a bug has been fixed that would result in the receiver code to become stuck on overrun.
• The terminals now support character attribute resetting in the VT200 style. This helps when using the terminals with current software that uses reverse video and assumes a VT200+ type terminal (like most Linux setups do).
• A bug has been fixed that caused the first character on a new line at the bottom of a scroll region to be replaced by a blank.

Cheers

Sytse

[Mark Matlock]

Sytse,

   This is very exciting news!! Congrats on all the developments!

    I still have one PiDP-11 left unassembled that I've been wanting to put together and I think this one should

have your FPGA technology in it. I'm particularly interested in the MINC emulation to compare with my

real MINC that is now using a QBone as its disk drive subsystem.

   The additional sensors for a weather station are also very appealing in your MINC configuration. One

question I had is who is a good vendor for the CYC1000 and the SDcard adapter board? Also, I'd be

interested in one of your shim boards if you have any left?

   Lastly, thanks for the new info on making a RK05 disk image of the MINC Basic OS

( https://pdp2011.sytse.net/wordpress/howto/minc-rk05/ )

   I had played with it a bit from emulated RX02s on the QBone but could never get it moved to an emulated hard

disk. These instructions explain that very well and I hope to have that working soon now.

   One final comment, I used your instructions on getting Basic Plus 2 V2.7 working from the corrupted tape

image on bitsavers. Today, it is possible to find a good copy of Basic Plus 2 V2.7 at Johnny Billquist's

Mim system ( ftp://ki...@mim.update.uu.se/ ) so that the installation matches the DEC documentation.

Many of Johnny's packages are also available via his new very convenient RPM package installer for those

who are on HECnet.

   Thanks again for your great work!!

Bes,

Mark Matlock

[Johnny Billquist]

Just a small observation.
RPM works just fine for anyone with any network connectivity. HECnet is
not required.

And speaking of BASIC+2 V2.7, not only do I have a correct tape. I also
have some patches for some minor things, that I'd recommend people grab.
BASIC+2 should identify as V2.7C when installed.

Johnny

> <https://pdp2011>.sytse.net/ and there also you'll find a post

> describing the changes (there are quite a few). I'll cut and paste
> the list here for your convenience:
>
> Base system
>

> * The RH disk controller used to only work with RP06 images. It

> now also works with RP04/RP05, RM05, and RP07. Also, it supports
> the Setasi/Shelby RP12 RM06 layout of 64 sectors, 32 tracks that
> will work with RSX-11MPlus for disk images up to 2GByte(with
> thanks to Johnny Billquist for the info and help), and a
> synthetic 64 sector, 64 track drive that works with 2.11BSD. The
> controller still only supports one disk drive at a time though
> (and I don't think I'll fix that soon).

> * Images for the 'artificial' drive types (RM06, RP2G) can be

> created by a modified version of simh that supports the new
> drive geometry. This has been tested up to a size of 2Gb (larger
> will not currently work). I haven't published the required
> (relatively minor) changes yet, but will try to do so soon.

> * The serial link driver did not work very well if it would

> receive short stop bits from the other end - potentially leading
> to garbled characters. It should now be lots more stable and
> capable of higher speed. The issue was analyzed by Steve Falco
> (thanks!).
>
>
> MINC
>

> * MINC now supports up to 64 A/D channels; the A/D channels

> present on the board are the lowest numbered channels, and all
> higher numbered channels are available to insert values
> into MINC that are obtained from other sources like sensors,
> other types of A/D converter, computed data, etc - as long as
> they more or less fit in a 12-bit signed integer.

> * All MNC module types can be configured to the maximum the

> software can handle: 1 MNCAD, 2 MNCKW, 4 MNCAA, 4 MNCDI and 4
> MNCDO. That means 16 DA channels, and 64 digital input and
> output channels - plus some control signals.

> * The logic in MNCDI for detecting state changes has been

> corrected; the previous version would only detect rising edge
> changes.

> * A driver for the PMOD DA4 (Analog Devices AD5628-1) has been

> added, next to the PMOD DA2 that was already available.

> * A driver for the PMOD NAV (Microchips LSM9DS1 and LPS25HB). The

> digital barometer in LPS25HB will likely feature in a future
> project for a weather station.

> * A driver for the PMOD HYGRO (TI HDC1080). The temperature and

> humidity sensors will also be useful for a future weather station.

> * A driver for the PMOD COLOR (AMS TCS3472). This sensor allows to

> determine the color temperature, and hence elegantly establish
> if the sun is shining or if there are clouds.

> * A driver for the ADXL345 accelerometer that is present on the
> DE0-Nano boards.
>
>
> VT100/VT105
>
> * The terminals now support double width and double height characters.
> * Automatic XOFF/XON has been added, and a bug has been fixed that

> would result in the receiver code to become stuck on overrun.

> * The terminals now support character attribute resetting in the

> VT200 style. This helps when using the terminals with current
> software that uses reverse video and assumes a VT200+
> type terminal (like most Linux setups do).

> * A bug has been fixed that caused the first character on a new

> line at the bottom of a scroll region to be replaced by a blank.
>
>
>
>
> Cheers
> Sytse
>

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com
> <mailto:pidp-11+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pidp-11/edddadc7-bf9b-4439-b24a-7c814f5473b4n%40googlegroups.com
> <https://groups.google.com/d/msgid/pidp-11/edddadc7-bf9b-4439-b24a-7c814f5473b4n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol

[Sytse van Slooten]

Hi Mark

On 31 May 2021, at 17:08, Mark Matlock <ma...@matlockfamily.com> wrote:

Sytse,

   This is very exciting news!! Congrats on all the developments!

    I still have one PiDP-11 left unassembled that I've been wanting to put together and I think this one should

have your FPGA technology in it. I'm particularly interested in the MINC emulation to compare with my

real MINC that is now using a QBone as its disk drive subsystem.

That would be most helpful - there are some details that I'm really not sure I have done right, and I have no way of verifying against a 'real' MINC myself. If you like, I'll send you some test cases (but not right away, it'll take some time to write up the questions and make sure the test cases are correct... not sure when I'll get around to that.

   The additional sensors for a weather station are also very appealing in your MINC configuration. One

question I had is who is a good vendor for the CYC1000 and the SDcard adapter board? Also, I'd be

interested in one of your shim boards if you have any left?

I'm not sure who carries cyc1000 in the US - apart from Arrow, of course. I get them directly from Trenz in Germany; as far as I understand Trenz designed the board with Arrow doing the distribution. But I don't use cyc1000 for the MINC systems myself, I use the de0nano and de10lite boards. They have a lot more I/O pins, have a/d capability. And de10lite also has a vga connector, and a much larger fpga. Mostly I use the de10lite for running MINC+VT105, and de0nano for MINC over a serial link to my laptop (and thus missing out on all the nice graphics...). Easier for development, at least for the sensors I've been working on.

In either case, I'm not sure about the pidp-11 + minc combo. Historically it isn't impossible, there are several references to 11/70, 11/34 etc in the MINC Basic executable - maybe DEC used those systems for development. But the blinkenlight pattern isn't very nice, it's similar to most RT-11 based systems - all 16-bit address and data lines on almost all the time, that's basically it. I usually don't include the console logic in my builds, disabling that speeds up the compilation by several minutes...

I have been thinking (well maybe dreaming is more correct) about making a front panel for MINC though - not as a replica of the real thing, but something like how DEC could've done it for a next generation 10 or 20 years later. Something with a few BNC connectors for the clock module Schmitt triggers, and lots and lots of blinkenlights for the MNCDI/MNCDO (and maybe even some way to actually use those signals as well, even though it is probably just an excuse for the blinkenlights). Maybe something like a backplane and modules, even. I'm not yet sure about what the front panel should be made of though, or how it should work with the modules. Or how sensor modules feeding into the A/D would fit in.

   Lastly, thanks for the new info on making a RK05 disk image of the MINC Basic OS

( https://pdp2011.sytse.net/wordpress/howto/minc-rk05/ )

   I had played with it a bit from emulated RX02s on the QBone but could never get it moved to an emulated hard

disk. These instructions explain that very well and I hope to have that working soon now.

Credit where it is due - most of the tricky bits there have been found by Jan Secker.

   One final comment, I used your instructions on getting Basic Plus 2 V2.7 working from the corrupted tape

image on bitsavers. Today, it is possible to find a good copy of Basic Plus 2 V2.7 at Johnny Billquist's

Mim system ( ftp://ki...@mim.update.uu.se/ ) so that the installation matches the DEC documentation.

Many of Johnny's packages are also available via his new very convenient RPM package installer for those

who are on HECnet.

Ah cool - I hadn't seen that! My RSX walkthru is getting really outdated (and my RSX-fu is not very strong either...). I should definitely redo it, and make an image that has Johnny's TCP on it as well.

   Thanks again for your great work!!

Bes,

Mark Matlock

Cheers!

Sytse

On Sunday, May 30, 2021 at 4:07:35 PM UTC-5 Sy...@sytse.net wrote:

Hi all,

I've just posted a new release of PDP2011 on https://pdp2011.sytse.net/ and there also you'll find a post describing the changes (there are quite a few). I'll cut and paste the list here for your convenience:

Base system

• The RH disk controller used to only work with RP06 images. It now also works with RP04/RP05, RM05, and RP07. Also, it supports the Setasi/Shelby RP12 RM06 layout of 64 sectors, 32 tracks that will work with RSX-11MPlus for disk images up to 2GByte (with thanks to Johnny Billquist for the info and help), and a synthetic 64 sector, 64 track drive that works with 2.11BSD. The controller still only supports one disk drive at a time though (and I don't think I'll fix that soon).
• Images for the 'artificial' drive types (RM06, RP2G) can be created by a modified version of simh that supports the new drive geometry. This has been tested up to a size of 2Gb (larger will not currently work). I haven't published the required (relatively minor) changes yet, but will try to do so soon.
• The serial link driver did not work very well if it would receive short stop bits from the other end - potentially leading to garbled characters. It should now be lots more stable and capable of higher speed. The issue was analyzed by Steve Falco (thanks!).

MINC

• MINC now supports up to 64 A/D channels; the A/D channels present on the board are the lowest numbered channels, and all higher numbered channels are available to insert values into MINC that are obtained from other sources like sensors, other types of A/D converter, computed data, etc - as long as they more or less fit in a 12-bit signed integer.
• All MNC module types can be configured to the maximum the software can handle: 1 MNCAD, 2 MNCKW, 4 MNCAA, 4 MNCDI and 4 MNCDO. That means 16 DA channels, and 64 digital input and output channels - plus some control signals.
• The logic in MNCDI for detecting state changes has been corrected; the previous version would only detect rising edge changes.
• A driver for the PMOD DA4 (Analog Devices AD5628-1) has been added, next to the PMOD DA2 that was already available.
• A driver for the PMOD NAV (Microchips LSM9DS1 and LPS25HB). The digital barometer in LPS25HB will likely feature in a future project for a weather station.
• A driver for the PMOD HYGRO (TI HDC1080). The temperature and humidity sensors will also be useful for a future weather station.
• A driver for the PMOD COLOR (AMS TCS3472). This sensor allows to determine the color temperature, and hence elegantly establish if the sun is shining or if there are clouds. 
• A driver for the ADXL345 accelerometer that is present on the DE0-Nano boards.

VT100/VT105

• The terminals now support double width and double height characters.
• Automatic XOFF/XON has been added, and a bug has been fixed that would result in the receiver code to become stuck on overrun.
• The terminals now support character attribute resetting in the VT200 style. This helps when using the terminals with current software that uses reverse video and assumes a VT200+ type terminal (like most Linux setups do).
• A bug has been fixed that caused the first character on a new line at the bottom of a scroll region to be replaced by a blank.

Cheers

Sytse

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/edddadc7-bf9b-4439-b24a-7c814f5473b4n%40googlegroups.com.

[Mark Matlock]

Sytse,

    Thank you for the additional information on the best FPGA to use for the MINC project. Your mention of the DE10Lite

reminded me that I may have one somewhere in the "Hobby Barn" and in fact I just found it. I had bought it for a totally

different project that I never got started back when I was working full time. Now that I'm retired I have a bit more

time but not as much as I had hoped.

   I'll start with the DE10Lite to hopefully reproduce your work on the MINC system. That will be interesting to compare

with my real MINC. Also, I'd be glad to test any questions you have on the real MINC. The MINC RT-11 Basic hides

much of the machine level interaction with the MINC modules and although they are quite similar to the earlier

LPS-11 systems that interfaced with Unibus PDP-11s (and would be a better fit to the 11/70) there are some

differences. I do have a LPS-11 in my collection but it is far from operational at this point.

   I would like to at some point build one of my unassembled PiDP-11s with one of the CYC1000 FPGAs but at this point

the MINC emulation is more intriguing. Also, I wanted to extend my thanks to Jan Secker for his work on moving the

MINC RT-11 Basic to a hard disk. The Qbone fits great in the space where the MINC normally has its I/O modules

and it gives great flexibility in moving files from the internet to the RT-11 environment via the BeagleBone's ftp. Also,

being able to boot RT-11 V5.7 w/ FEP/FRG or the RT-11 V3 MINC Basic or even RSX11M (where I wish I could find a copy of the

RSX FEP/FRG software like the RT-11 FEP/FRG package.

  The front panel of a MINC replica of the future, now past is an interesting concept. Before the QBone I used an

Emulex UC07 and a SCSI2SD card as a disk drive. I burned eight 2716 EPROMs with the firmware from a PDP-1123+

and put them on the BDV11 board so that I had an easy way to boot MSCP drives. I also wanted the SDcard reader

on the front the MINC along with a drive activity LED. I made a replicate blank panel in epoxy in a silicon mold (this

was when 3D printing was still pretty new). Then I milled a slot and glued a SDcard reader to micro sdcard cable

to go in the V5 SCSI2SD card. It has worked fine, but it is tricky to move a few files to a RT-11 disk image then

copy that to partition on the SDcard without making a mistake.

  The DE10Lite has the digit display so maybe the replica should add the LED display from the LPS-11 to the new

MINC repllica concept. Perhaps a small OLED display could even add back the oscilloscope functionality of the

original DEC LINC systems? LEDs to display the Digital Output states would be another easy add with bar graph

segment LEDs, Fun to think about this.

Thanks,

Mark

[John H. Reinhardt]

Originally sent to Sytse directly - Oops. Meant to go to the group/

On 5/31/2021 5:23 PM, Sytse van Slooten wrote:
> Hi Mark
>

>> On 31 May 2021, at 17:08, Mark Matlock <ma...@matlockfamily.com <mailto:ma...@matlockfamily.com>> wrote:
>>
>>
>> Sytse,
>>    This is very exciting news!! Congrats on all the developments!
>>
>>     I still have one PiDP-11 left unassembled that I've been wanting to put together and I think this one should
>> have your FPGA technology in it. I'm particularly interested in the MINC emulation to compare with my
>> real MINC that is now using a QBone as its disk drive subsystem.
> That would be most helpful - there are some details that I'm really not sure I have done right, and I have no way of verifying against a 'real' MINC myself. If you like, I'll send you some test cases (but not right away, it'll take some time to write up the questions and make sure the test cases are correct... not sure when I'll get around to that.
>
>>
>>    The additional sensors for a weather station are also very appealing in your MINC configuration. One
>> question I had is who is a good vendor for the CYC1000 and the SDcard adapter board? Also, I'd be
>> interested in one of your shim boards if you have any left?
> I'm not sure who carries cyc1000 in the US - apart from Arrow, of course. I get them directly from Trenz in Germany; as far as I understand Trenz designed the board with Arrow doing the distribution. But I don't use cyc1000 for the MINC systems myself, I use the de0nano and de10lite boards. They have a lot more I/O pins, have a/d capability. And de10lite also has a vga connector, and a much larger fpga. Mostly I use the de10lite for running MINC+VT105, and de0nano for MINC over a serial link to my laptop (and thus missing out on all the nice graphics...). Easier for development, at least for the sensors I've been working on.
>
>

Just as an update on the CYC1000.  When this message was first posted on 31-May, Arrow had 0 stock on the CYC1000.  Trenz was the same.  But checking today I see Arrow has 143 in stock and shippable. What's interesting is that the price has gone up by about $20 (US) to $57.95.  I paid $35.10 back in Nov 2019.  I see Trenz is still saying 0 stock on their site.  Google finds no others with any stock listing.


--
John H. Reinhardt

[John H. Reinhardt]

And I just looked again today and they have none!  Back to 0 in stock.  That 143 sold fast.  I wonder who is using them.

--
John H. Reinhardt

[atreu...@gmail.com]

Syste,

I have been a bit lost in trying to build a RSX11 system from Johnnie's latest drop of RSX11.  I have a CYC1000 and shim that runs fine as an RP06.  But I am trying to build a bigger system and can't figure out the simh disk config to sysgen and get to boot on the CYC1000 

Aug Treubig

On Sunday, May 30, 2021 at 4:07:35 PM UTC-5 Sy...@sytse.net wrote:

[Sytse van Slooten]

Hi Aug

Are you trying to get the 2Gbyte disk drives to work? That requires source patches to simh to add that drive type - there's no support in the official versions. 

Tbh I've not looked at it at all since the release, I'd have to dig out if I still have the patch somewhere and against which version of simh it is.

RP07 is much easier, are you sure that's not big enough? About half a gig :-)

Cheers

Sytse

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/160af17b-2299-4c31-a2ca-a91cd3ebf33an%40googlegroups.com.

[atreu...@gmail.com]

Hi Sytse,

Half gig would be fine.  So i will look at the RP07.   

Thanks

Aug

[August Treubig]

Ok, 

I am doing something wrong.   

Hello, world [t47]: cpu 11/70 fpu

177776           psw

177774           slr

177772           pir

177770           mbr

177766           cer

177764           id

177762 - 177760  size

177752           mr

177750           mnt

177746 - 177740  mr

177676 - 177600  pdr/par

177576 - 177572  mmu

177570           sdr

177566 - 177560  kl

177546           kw

176776 - 176700  rh

174516 - 174510  xu

173776 - 173000  m9312

172516           mmu

172376 - 172300  pdr/par

172276 - 172200  pdr/par

170376 - 170200  ubm

165776 - 165000  m9312

boot from rp:

SAV -- Booted device not in system - DB 176700

Aug

Sent from my iPad

On May 13, 2023, at 11:55 AM, atreu...@gmail.com <atreu...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/d8415505-1f95-453d-aeef-fa7558c6f878n%40googlegroups.com.

[Johnny Billquist]

I think that message would be from SAV if the device driver don't exist,
but I might be remembering wrong...?

Johnny

[August Treubig]

Ok, I figured it out. Is a vhdl config thing in top.vhd.

have_rh => 1,
rh_type => 7, <=====. Added this to set to RP07 instead of default RP06.
rh_sdcard_cs => sdcard_cs,
rh_sdcard_miso => sdcard_miso,
rh_sdcard_mosi => sdcard_mosi,
rh_sdcard_sclk => sdcard_sclk,
rh_sdcard_debug => sddebug,


Aug


> On May 14, 2023, at 7:33 AM, Johnny Billquist <b...@softjar.se> wrote:

> --
> You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/0267f5f7-e541-b58d-e618-4d8757d575a2%40softjar.se.

[Crni Mrki]

Hmmmm...

I have problems with RP07 too:

Hello, world [t47]: cpu 11/70 fpu
177776           psw
177774           slr
177772           pir
177770           mbr
177766           cer
177764           id
177762 - 177760  size
177752           mr
177750           mnt
177746 - 177740  mr
177676 - 177600  pdr/par
177576 - 177572  mmu
177570           sdr
177566 - 177560  kl
177546           kw
176776 - 176700  rh
174516 - 174510  xu

173776 - 173000  m9312165776 - 165000  m9312

boot from rp:

SAV -- Booted device not in system - DR 176700

172516           mmu
172376 - 172300  pdr/par
172276 - 172200  pdr/par
170376 - 170200  ubm

pdp11-20210530 version:

      have_rh => 1,
      rh_type => 7,

      rh_sdcard_cs => sdcard_cs,
      rh_sdcard_miso => sdcard_miso,
      rh_sdcard_mosi => sdcard_mosi,
      rh_sdcard_sclk => sdcard_sclk,
      rh_sdcard_debug => sddebug,

Genned system booted in simh:

RSX-11M-PLUS V4.6  BL87   1024.KW  System:"RSXMPL"

...

...

>dev
VF0:  Offline Loaded Type=unknown
VF1:  Offline Loaded Type=unknown
CO0:  TT0: Loaded
TT0:  [200,200]   [200,200] - Logged in  Loaded
VT0:  Loaded
VT1:  Loaded
RD0:  Loaded
DR0:  Public Mounted Loaded Label=RSX11MPBL87 Type=RP07  
 Seek_Optimization=Nearest:10. Overlapped_Seeks

DU0:  Loaded Type=RD54  
MS0:  Offline Loaded Type=unknown
NL0:  Loaded
TI0:
CL0:  TT0:
SP0:  DR0:
LB0:  DR0:
SY0:  DR0:
>

What is wrong? Boot in simh, there is DR driver, rh_type is 7 in top.vhd.

[Crni Mrki]

To answer to myself again :-)

Problem was in sd card.
After several times, after I transfer the image to the card with DD, the phenomenon occurs that DD completes the job 

but the transfer is not successful and the old image remains on the card. I replaced the card and now everything is OK.
Thank you all for your help :-) .