Hi I applied both patches to kernel 2.6.11. It didn't seem to make a lot of difference. I've attached the tzap output. Ben I might have posted twice ! Before [strider@localhost ~]$ tzap -r "SBS DIGITAL 1" using '/dev/dvb/adapter0/frontend0' and '/dev/dvb/adapter0/demux0' tuning to 571500000 Hz status 1f | signal ffff | snr 9789 | ber 000003d6 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7da7 | ber 00002175 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7be7 | ber 0000218d | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7e77 | ber 00002017 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 8268 | ber 00002093 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7f27 | ber 00001fd8 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7e37 | ber 00002025 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 78c7 | ber 000018bc | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 79e7 | ber 000017a4 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7bb7 | ber 00001756 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7a57 | ber 000017d1 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 79e7 | ber 00001417 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7b07 | ber 0000144c | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7967 | ber 00001a96 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 78c7 | ber 000017f2 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7a97 | ber 00001593 | unc 00000000 | FE_HAS_LOCK [1]+ Stopped tzap -r "SBS DIGITAL 1" [strider@localhost ~]$ After [strider@localhost ~]$ tzap -r "SBS DIGITAL 1" using '/dev/dvb/adapter0/frontend0' and '/dev/dvb/adapter0/demux0' tuning to 571500000 Hz status 1f | signal ffff | snr 7ea7 | ber 00001822 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 8098 | ber 00001efd | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7fd7 | ber 00001cfe | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 80e8 | ber 00001bfa | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7e67 | ber 00002012 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7f47 | ber 00001c98 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7f87 | ber 00001e67 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 79e7 | ber 000013d3 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7af7 | ber 000014af | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7db7 | ber 000012d6 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7c07 | ber 00001230 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7e87 | ber 00001060 | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7de7 | ber 000015ff | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7c07 | ber 0000132a | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7d47 | ber 0000154a | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7ee7 | ber 0000177b | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7cc7 | ber 0000122d | unc 00000000 | FE_HAS_LOCK status 1f | signal ffff | snr 7d67 | ber 00001436 | unc 00000000 | FE_HAS_LOCK [1]+ Stopped tzap -r "SBS DIGITAL 1" [strider@localhost ~]$ > Jyrki Niskala <jyrki.n@xxxxxxxxx> wrote: > > Johannes Stezenbach wrote: > > >On Sat, Sep 03, 2005 Jyrki Niskala wrote: > > > > > >>This patch gives better performance when using high frequencies (over > >>470 MHz). I got those new > >>charge pump values by i2c-snooping windows drivers. > >> > >> > >... > > > > > >>--- dvb-kernel/linux/drivers/media/dvb/bt8xx/org-dvb-bt8xx.c > >>2005-09-03 19:21:55.000000000 +0200 > >>+++ dvb-kernel/linux/drivers/media/dvb/bt8xx/dvb-bt8xx.c > >>2005-09-03 19:23:11.000000000 +0200 > >>@@ -264,10 +264,10 @@ > >> cpump = 1; > >> else if (params->frequency < 470000000) > >> cpump = 2; > >>- else if (params->frequency < 750000000) > >>- cpump = 2; > >>- else > >>- cpump = 3; > >>+ else if (params->frequency < 750000000) > >>+ cpump = 0x40; > >>+ else > >>+ cpump = 0xc0; > >> > >> if (params->frequency < 175000000) > >> band_select = 0x0e; > >> > >> > > > >Hm, that's not the real fix, there seems to be another bug. > >Can you try the following patch: > > > > > >--- linux/drivers/media/dvb/bt8xx/dvb-bt8xx.c 10 Sep 2005 07:26:02 > -0000 1.49 > >+++ linux/drivers/media/dvb/bt8xx/dvb-bt8xx.c 10 Sep 2005 16:52:52 > -0000 > >@@ -280,7 +280,7 @@ static int microtune_mt7202dtf_pll_set(s > > data[0] = (div >> 8) & 0x7f; > > data[1] = div & 0xff; > > data[2] = ((div >> 10) & 0x60) | cfg; > >- data[3] = cpump | band_select; > >+ data[3] = (cpump << 6) | band_select; > > > > i2c_transfer(card->i2c_adapter, &msg, 1); > > return (div * 166666 - 36000000); > > > >Additionally the Windows driver seems to use slightly > >different charge pump settings (1 vs. 2 in the 470...750MHz range), > >can you test which one actually works better? > > > >Thanks, > >Johannes > > > > > > > I think your patch affects all frequencies, i have only 4 mux from 538 > MHz to 778 MHz to test on. > But I will give it a go asap. > > By the way, this is my i2c log, it may be a outdated windows driver > since the log is from april 2004 ;-) > > Milksop GPL Reflasher - 0.20 - (c)2002 andy@xxxxxxxxxxx > ... > ### zap to TV3 @ 706 MHz: > 405uS: 70 W 02 06 00 01 . . . > 690uS: 60 W 11 64 88 43 d . C > 627uS: 70 W 0C 18 . > 356uS: 70 r 00 0D . > 433uS: 70 W 02 06 00 00 . . . > ### zap to TV4 @ 746 MHz: > 432uS: 70 W 02 06 00 01 . . . > 636uS: 60 W 12 54 88 43 T . C > 612uS: 70 W 0C 18 . > 374uS: 70 r 00 0D . > 463uS: 70 W 02 06 00 00 . . . > ### zap to SVT1 @ 778 MHz: > 400uS: 70 W 02 06 00 01 . . . > 606uS: 60 W 13 14 88 C3 . . . > 605uS: 70 W 0C 18 . > 378uS: 70 r 00 0D . > 397uS: 70 W 02 06 00 00 . . . > #### zap to ZTV @ 538 MHz: > 403uS: 70 W 02 06 00 01 . . . > 646uS: 60 W 0D 74 88 43 t . C > 629uS: 70 W 0C 18 . > 358uS: 70 r 00 0D . > 422uS: 70 W 02 06 00 00 . . . > ... > > / Jyrki > > > > _______________________________________________ > > linux-dvb@xxxxxxxxxxx > http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
