Topics

SkyStar card versions - calibration differences?

David J Taylor GM8ARV 🏴󠁧󠁢󠁳󠁣󠁴󠁿 🇪🇺
 

Having just moved from a V2.3 hardware with V2.4.2 (?) software to a
V2.6B card with the V4.2.8 drivers, I have found that, as others
have reported, the signal level indications are completely different!

What was 60% "Signal Quality" has become 80% (perhaps), what was
9.3dB "SNR" has become 14.5dB (perhaps). Is the new card that much
better, or has someone changed the calibration levels? Is the
calibration that arbitrary? What does "Signal Quality" mean in any
case?

(I should add that I also now have a spliiter so that I can feed two
cards from one antenna. My simplistic view suggests that splitting
the signal should, er, decrease the SNR and not increase it!).

If nothing else, it does mean that when reporting levels you must
specify what hardware and software you are using! Comments
welcomed...

Cheers,
David

James Brown
 

In message <bojnnb+mu9s@...>, David Taylor <david-taylor@...> writes
Having just moved from a V2.3 hardware with V2.4.2 (?) software to a
V2.6B card with the V4.2.8 drivers, I have found that, as others
have reported, the signal level indications are completely different!

What was 60% "Signal Quality" has become 80% (perhaps), what was
9.3dB "SNR" has become 14.5dB (perhaps). Is the new card that much
better, or has someone changed the calibration levels? Is the
calibration that arbitrary? What does "Signal Quality" mean in any
case?

(I should add that I also now have a spliiter so that I can feed two
cards from one antenna. My simplistic view suggests that splitting
the signal should, er, decrease the SNR and not increase it!).

If nothing else, it does mean that when reporting levels you must
specify what hardware and software you are using! Comments
welcomed...

Cheers,
David
Was that before or after your aerial installer readjusted your dish David?

James
--
James Brown

David J Taylor GM8ARV 🏴󠁧󠁢󠁳󠁣󠁴󠁿 🇪🇺
 

Was that before or after your aerial installer readjusted your
dish, David?

Fair point, James. The dish alignment turned the varying signal
level bottoming at 60% into a varying signal level peaking at 70%.
Why the variation was still there I don't know. The SNR increased
from just over 9dB to around 10dB, with fluctuations.

The difference between the cards and their associated software is
far greater (which is why I suspect the difference is not as marked
as the screen read-out would suggest).

Cheers,
David

a_van_belle
 

David,

The skystar driver shipped May 2003 was 4.2.2

The new rev 2.6B Skystar does readout SQ a bit lower normally.
But as you have noticed, using a 2 way splitter does change SQ
readings.
I did expect it to decrease but to my surprise it increased overhere
from 61 to 68%. And was depending on the fact if and what type
receiver was connected to the second port.

Only explanation: Either the splitter atennuates noise more than
signal, giving a better SNR or the receiver was "oversteered" in the
first place.

Only way to find out is to place a atennuator and see what happens.

For more info on the 2 way splitter (the same type is marketed under
different brandnames) can be found on:

http://www.triax.dk/ifs/files/triax/is/presentation/home/Satellite_x_t
errestrial/DTH_(Direct_to_Home)/Receivers/FB/5-2400_MHz/TSS-
SAT_Splitter/TSS-SAT_Splitter.jsp

But what really counts, do you have less missing segments ?
Today I only have missing segments:
1345 lrit ch01 + ch03 seg4
1445 hrit ch01 seg 4 + ch12 seg 10+11
Could be caused by a PC problem, I did notice abnormal harddisc
activity around these times. Guess it's time to do some defragging !

Greetings,
Arne van Belle

David J Taylor GM8ARV 🏴󠁧󠁢󠁳󠁣󠁴󠁿 🇪🇺
 

--- In MSG-1@..., "a_van_belle" <a.van.belle@h...> wrote:
David,

The skystar driver shipped May 2003 was 4.2.2
Thanks.

The new rev 2.6B Skystar does readout SQ a bit lower normally.
But as you have noticed, using a 2 way splitter does change SQ
readings. I did expect it to decrease but to my surprise it
increased overhere from 61 to 68%. And was depending on the fact if
and what type receiver was connected to the second port.

Only explanation: Either the splitter atennuates noise more than
signal, giving a better SNR or the receiver was "oversteered" in the
first place.

Only way to find out is to place a atennuator and see what happens.
We say "over-driven" but I knwo what you mean - too high a signal
level.

[]
But what really counts, do you have less missing segments ?
Today I only have missing segments:
1345 lrit ch01 + ch03 seg4
1445 hrit ch01 seg 4 + ch12 seg 10+11
Could be caused by a PC problem, I did notice abnormal harddisc
activity around these times. Guess it's time to do some defragging !

Arne, my own log shows:

Missing HRIT & LRIT segments report
Cycle: 1200
msg-ch11, seg: 3
Cycle: 1245
msg-ch04, seg: 3
lrit-ch04, seg: 3

and not the segments you missed. Around 12:45 segment 3 the testing
PC with the new DVB card was closed down, so there could have been a
transient on the shared LNB control line. What I have now done
effectively is to upgrade the receiver PC from a 550MHz PIII with
the V2.3 SkyStar card to a 1GHz PIII with the newer V2.6B card (and
also with a complete fresh install of Windows 2000 Prof SP4).

I do notice that with the new card, in either of the two PCs where
it has been tested, the mouse movement is subject to occasional
judder, gaps of motion lasting a fraction of a second. This judder
does not occur when the card is not installed and receiving data.
Both PCs have a PS2 mouse.

What I found with the testing PC (2.8GHz P4) was that whilst
most "office" use of the PC did not produce missing segments,
running anything that was very CPU intensive could result in
segments begin missed. E.g. making panoramas, and some Paint Shop
Pro filtering or resampling operations. Eating CPU for more than
(say) one second was not on. (On the other hand, running disk
defrags, anti-virus, Zone Alarm etc. seems to be OK. Perhaps
because they involve I/O as well as CPU activity). I had hoped to
be able to run the DVB card on a general "office" PC, but that
proved not to be 100% reliable.

Cheers,
David

a_van_belle
 

--- In MSG-1@..., "David Taylor" <david-taylor@b...>
wrote:
I do notice that with the new card, in either of the two PCs where
it has been tested, the mouse movement is subject to occasional
judder, gaps of motion lasting a fraction of a second. This judder
does not occur when the card is not installed and receiving data.
Both PCs have a PS2 mouse.
Could be a shared IRQ David. Did you try the Old card on the 4.2.8
driver too ?
Testing different hardware revisions with the newest driver and SP4
is quite a bold step David. But if it works, does prove that some
previous error reports on 4.2.8 and Microsoft patches might have been
just a unlucky coincidence.

What I found with the testing PC (2.8GHz P4) was that whilst
most "office" use of the PC did not produce missing segments,
running anything that was very CPU intensive could result in
segments begin missed. E.g. making panoramas, and some Paint Shop
Pro filtering or resampling operations. Eating CPU for more than
(say) one second was not on. (On the other hand, running disk
defrags, anti-virus, Zone Alarm etc. seems to be OK. Perhaps
because they involve I/O as well as CPU activity). I had hoped to
be able to run the DVB card on a general "office" PC, but that
proved not to be 100% reliable.
Monitoring CPU load is required, unfortunately I don't have a tool to
measure the current harddisc load. Both are very important.

By the way, I see on MSG animator that the clouds are just clearing
over here at this moment, so I might enjoy the moon eclips this
night !

Greetings,
Arne van Belle

David J Taylor GM8ARV 🏴󠁧󠁢󠁳󠁣󠁴󠁿 🇪🇺
 

Could be a shared IRQ David. Did you try the Old card on the 4.2.8
driver too ?
Mouse judder: DVB card IRQ 09, Mouse IRQ 12
Looking at "Resources by connection" shows that the mouse on (ISA)
IRQ 12 shares that with nothing, but the DVB card on (PCI) IRQ 9
seems to be charing with the sound card, the video card, the network
card and two USB controllers!

Interestingly, on a recent PC most of the PCI IRQ values are
different, but on the 1GHz PC all are at 9....

Testing different hardware revisions with the newest driver and
SP4 is quite a bold step David. But if it works, does prove that
some previous error reports on 4.2.8 and Microsoft patches might
have been just a unlucky coincidence.
Well, when you install the new card the drivers on the CD are
V2.4.8, so I presume that they may be required....

Monitoring CPU load is required, unfortunately I don't have a tool
to measure the current harddisc load. Both are very important.
It seems to me that it is something more like a timeout issue rather
than absolute CPU or disk load - if the DVB card doesn't serviced
within a certain interval data is lost. Of course, there is nothing
like enough buffering on the card. If there was 30 seconds, say, of
data buffer (about 8MB, say 16 or 32MB for good measure, I am sure
the card would be much robust to use in a general purpose PC.

By the way, I see on MSG animator that the clouds are just
clearing
over here at this moment, so I might enjoy the moon eclips this
night !
It was cloudy for us in Edinburgh.

Cheers,
David

Peter Benney <tugboat@...>
 

(I should add that I also now have a spliiter so that I can feed two
cards from one antenna. My simplistic view suggests that splitting
the signal should, er, decrease the SNR and not increase it!).
Today I added a splitter and fed the Skystar PCI card and a Digital Satellite Receiver(STB) from the same LNB.

The signal on the STB dropped from 88% to 86% and quality remained at 100% . Setup4PC showed no change in Quality 69-71% but with a slightly more rapid fluctuation. The SNR changed from a steady 10.8 to varying between 10.8 and 11.2. The insertion loss of the splitter is 4dB.

Setup4PC showed no change when connected either via the STB loopthrough or directly to the LNB. An advantage of the loopthrough setup is the ability to monitor a TV channel on the 11096 tp whilst receiving MSG1 data. The motorised dish allows the tweaking of the azimuth with the aid of the more responsive signal and quality indicators of the STB. So far since installation my dish has remained rock steady .

Peter