Search options (Page 47 of 99)

14 and 12 no longer merged
14 gone

after putting them in the sound_B excell sheet and looking at dupes, I have those folders left.
12 / 17 / 23 / 25 / 30

see you tomorrow

hi
30 folders, 15 was empty

work done so far was to find duplicates in those 29.
At the moment, folder 12 and 14 got a merge into 14.
(the smwdmCH5 INF in 14 was kept, the other INFs from 12 were added, and the 12's cat was renamed. The systemfiles are same.).

After this preliminary work, I have 11 folders left.
04 / 07 / 14 (with the added INFs and CAT from 12)  / 16 / 17 / 22 / 23 / 24 / 25 / 28 / 30

typo, I missed 25 in the 11 folders.

thanks.

Hi
I uploaded a mass with a change in Nvidia, but could not reach you.

I looked at that one too. It must have been an O for OLD, but the driver in there is the newer.
The changelog might shed a light.

@ debugger.
we can do just like last time.
(it will save connection time on the 56K modem..)
Upload the additions to FTP, and let us put them in the Pack for you.
We'll then put the Pack up for test (big upload), and later publish (another big upload).

hi,
yes, I worked about 14 hours on this one, because duplicate HWIDs cause trouble.
Oh, I have done some more checks after I uploaded, reran the tools I used to collect the HWIDs, and found that I had forgotten to delete a folder.
The corrected "beta" is uploading right now.
MD5 hash
0ddc45ca25c9e892abc528fafbe3a81b DP_Sound_B_wnt5_x86-32_7106.7z

========

These INF files were removed

S\A aliOEM8T removed
M\1 MaRevo.inf removed
SM\E (swwdmCH5.INF removed)
SM\F\1 (swwdmCH4.INF REMOVED)
S\T\5 (MXR6Fire.INF removed)

these drivers were removed

SM\D\1
SM\G\2
SM\G\4
SM\G\5
SM\G\6
SM\H\1
SM\Z\3
S\S   (driver in root removed)
S\S\T
T\1
T\5
----------------

INF "pruning".
A large number of duplicated HWIDs were removed.
For instance; CR\6 and CR\7  had huge overlap.
A Folder compare tool tells me that about fifty INF files have had HWIDs pruned from them.
Some drivers have more than one INF file, and a few of those INF files became totally obsolete (no unique HWIDs) so they were deleted.

Some drivers were completely covered by other newer drivers, and were removed.
I made notes, and also have my Excell worksheet.
A rough count tells me I pruned about 900 duplicated HWIDs out of the pack.
(about One Fifth of the total HWIDs, that is.)

Let's hope this work was not in vain.

hi Dolivas,

fixing this will not be easy.
We've found duplicated generic HWIDS across ALi, C-media, sigmatel and analog devices.
(different chips for which one finds drivers with SAME manufactor generic HWID.. aargh.)

I'll start work on this after acronis finishes cloning my systemdrive(s*) in the new rig.
(* the intel RAID tools told me one of the RAID member drives has an error. I had to do a surface scan/fix before cloning could be succesful.)

I already sent excell worksheets to team members, so they also know where the problems are.
We want to avoid duplicated work too, but at the moment I cannot use my machine to get in contact.
Anyway, we will look into it.

I cannot work on any packs tonight.
I have to migrate my hard disks into another machine.
(a backup was made, twice..)

Hi
well, I simply turn ON QSC.

I just saw you spotted a problem in SOUND forum, and I looked at an excell file only moments ago.
(It's a file with all the HWIDS from SOUND-B in which the duplicates are hilited..)

You will want to see the problem, I guess.
It's a small wonder some drivers still work.

Dupes in same driver have to be filtered out before the excell file is really usable.
However, that is fairly easy
The target OS is one cause for dupes.
Then, for instance, there can be more than one INF file in same same driver, and the folder location/driver version tells us which are appearing more than once for same driver.

After that, it gets to be like REAL work.
If the device is same, OKAY, I think it can be left as is, but...

BUT, I found  several HWID duplicated for analog devices, for C-media and for sigmatel..
If you want the tool to build these Excell files, shoot me an email.

OH, maybe you can turn QSC ON for a test.
If you get no device driver signing problem after that, you could file a bug report.

Hi
Ever since the BartPE (ubcd4win) plugin was introduced, I have seen the QSC lines appear twice IF QSC is turned OFF.
(I think that the recommendation to not use it for method 2 should be dropped, because when it is turned ON for M2 there are few failure reports. And for Method 1, the little bugs become a BIG one, because slipstream fails.)

Oh, before the plugin we had Base 7.04, and I told OverFlow 7.04 already had an issue when QSC was turned OFF and that Base 7.01 has it too.
Some people report missing files if QSC is NOT used.

Hi

I'll do this asap.