The Nextor Work-In-Progress topic

Page 6/8
1 | 2 | 3 | 4 | 5 | | 7 | 8

By rjp

Master (193)

rjp's picture

17-04-2014, 21:37

I started to give some order to chaos, and prepare some Compact Flash cards to use with my MSXs. And I found a annoying problem: Lack of slack space. There is 1,3 Gb of MSX-related files in the CF card, and Nextor told me that it was full.

Is there any way to solve this, maybe increasing the number of sectors, or decrease the sector size? Maybe the definitive solution is to use FAT32, but it's not the proper time. =)

But DSK association to a drive letter can be a way to avoid this problem... Hmmm, I need to try the newest Nextor release.

By Prodatron

Paragon (1801)

Prodatron's picture

17-04-2014, 21:50

1,3GB usable space on 2GB or 4GB fat16? Both sounds realistic, as most MSX files are much smaller than 64K or 32K, but each single file occupies that space. Yes, FAT32 would allow much smaller cluster sizes, so you would use the disc space much more efficiency. But I wonder if this is really necessary for retro software? I was always sure, that such partition sizes are more usefull for MP3s, videos etc...

By rjp

Master (193)

rjp's picture

19-04-2014, 15:18

Hmmm, I can use DSK images as a workaround, mounting them as drive letters. It'd be funny, using a fs into another fs... Today, it's the way that I can think about decreasing slack space.

Well, I need to flash Nextor 2.1 alpha 1 in my Tecnobytes's IDE Mapper. Here we go.

BTW, is there any solution for the "long name problem"? Most of the files which are in the CF cards has more than 8 letters, and renaming all of them would be a annoying task.

By sd_snatcher

Prophet (3486)

sd_snatcher's picture

19-04-2014, 17:53

Either that or you can use a bit smaller partitions. That will reduce the cluster size to a more sane value and consequently reduce the wasted space. A 2GB partition will have 32KB clusters, while 1GB partitions will have 16KB clusters.

Most MSX1 ROMs have 16KB and will occupy twice that on a 2GB FAT16 partition.

By rjp

Master (193)

rjp's picture

26-04-2014, 06:13

Yeah, I prepared a 4-partition CF card, where each partition has 1 Gb. Wow, slack space was greatly decreased. Maybe it's the way until we don't have FAT32 support. :-)

By rjp

Master (193)

rjp's picture

26-04-2014, 21:08

BTW, using mapdrv to assign partitions to drive letters was a bit tricky, and the Getting Started manual wasn't very clear about this. If I may talk, I think it would be better if it's better explained.

By konamiman

Paragon (1157)

konamiman's picture

28-04-2014, 15:54

rjp wrote:

BTW, using mapdrv to assign partitions to drive letters was a bit tricky, and the Getting Started manual wasn't very clear about this. If I may talk, I think it would be better if it's better explained.

I'm sorry to read that. Any suggestion on how to improve the documentation will be welcome, as I can't figure out by myself how to do it. Smile

By konamiman

Paragon (1157)

konamiman's picture

28-04-2014, 16:04

rjp wrote:

Why not RALLOC patches Nextor, in order to report 0 bytes, and doesn't make any calculations? It would be very interesting.

(The quote is from a comment in the Nextor 2.1 Alpha 1 newspost)

Actually you are not the first person suggesting this...

The original purpose of the Reduced Allocation Information mode was to provide a free space amount information that does never exceed 32MB, to prevent old applications that expect the (cluster amount x sectors per cluster) figure to be within the 0-32MB range from choking on higher numbers. The reduction in the free space caltulation time is a (nice) collateral effect, but as I have been suggested, the time needed for calculating the free space on a heavily filled partition will not be significantly reduced even with RALLOC in place.

So I think that a "zero allocation information mode" is indeed a good idea, but not as a substitute for RALLOC, but rather as a complementary mode. I'll take a look at it.

By Grauw

Ascended (10181)

Grauw's picture

28-04-2014, 16:10

Won’t reporting 0 bytes free cause problems with software that checks the free space before writing a file?

I still think there must be some other way as both the ESE OCM and Padial SD do report free space rapidly, as mentioned in that thread…

By konamiman

Paragon (1157)

konamiman's picture

14-05-2014, 16:33

To follow up comments in the Nextor 2.0.3 release newspost:

edoz wrote:

Just performance problem on the sunrise (...) It seems when I write to the disk this takes a lot of time

Only with the Sunrise IDE interface? You are not experiencing the same problem with your Megaflash devices?

Could you provide me some more information? Please tell me:

- The capacity of the device where you are writing the file
- How much free space does it have
- The size of the file being copied
- How much time does it take the file write operation

Grauw wrote:

I think the speed is influenced by the cluster count, not the cluster size.

Indeed. The calculation time will be the same for 512M, 1G, 2G and 4G partitions, since all of them have a little less than 64K clusters.

Page 6/8
1 | 2 | 3 | 4 | 5 | | 7 | 8