hardrive and dvd drive with the same ide cable

Electro · 07-02-2009, 08:43 PM

The performance is normal for IDE when both a hard drive and an optical drive is connected to the same channel. Both drives are fighting for attention, so one drive have to be paused in order for the other to be used. This pause is a latency penalty and nothing having to do with throughput. The rule of thumb for IDE controllers is one device per channel or categorize each channel for either hard drive only or optical drive only. If all you have is a one channel IDE controller, you can either add an IDE controller or use SATA controller that I assume you have. You can use SATA to IDE converter, but you have to edit /etc/fstab for the drive changes. To make the transition easier is to set labels for each partitions and mount by labels. If you do change things around, Linux will always boot up in a reliable and predictable manner when using labels. Sure UUID could be used but not all file systems support it.

Hard drives barely reaches 66 megabytes per second. They average around 30 megabytes per second. UDMA-33 or UDMA-66 is plenty. All the need and want for UDMA-133 or SATA 300 for hard drives is just about stupidity.

GrapefruiTgirl · 07-02-2009, 09:14 PM

Quote:

Originally Posted by Electro

..Hard drives barely reaches 66 megabytes per second. They average around 30 megabytes per second. UDMA-33 or UDMA-66 is plenty. All the need and want for UDMA-133 or SATA 300 for hard drives is just about stupidity.

While this was entirely true a few years ago (not the 'stupidity' part), and is still often true for mobile HDD's and lower end or older desktop HDD's, it soon will not be true at all.

Look here for a tiny idea of what some HDD's can do:
http://www.techarp.com/showarticle.a...tno=511&pgno=3

Also, for everyone's reference (including my own):

Quote:

Originally Posted by http://en.wikipedia.org/wiki/Parallel_ATA

"Lowest speed"

It is a common misconception that, if two devices of different speed capabilities are on the same cable, both devices' data transfers will be constrained to the speed of the slower device.

For all modern ATA host adapters this is not true, as modern ATA host adapters support independent device timing. This allows each device on the cable to transfer data at its own best speed. Even with older adapters without independent timing, this effect only applies to the data transfer phase of a read or write operation. This is usually the shortest part of a complete read or write operation.

"One operation at a time"

This is caused by the omission of both overlapped and queued feature sets from most parallel ATA products. Only one device on a cable can perform a read or write operation at one time, therefore a fast device on the same cable as a slow device under heavy use will find it has to wait for the slow device to complete its task first.

However, most modern devices will report write operations as complete once the data is stored in its onboard cache memory, before the data is written to the (slow) magnetic storage. This allows commands to be sent to the other device on the cable, reducing the impact of the "one operation at a time" limit.

The impact of this on a system's performance depends on the application. For example, when copying data from an optical drive to a hard drive (such as during software installation), this effect probably doesn't matter: Such jobs are necessarily limited by the speed of the optical drive no matter where it is. But if the hard drive in question is also expected to provide good throughput for other tasks at the same time, it probably should not be on the same cable as the optical drive.

jay73 · 07-02-2009, 09:52 PM

Please please I want cheaper SSDs now

golgo · 07-02-2009, 10:28 PM

thank you...well i aready have it as a master and it still the same...so if i change the cable it wont go faster right??

GrapefruiTgirl · 07-02-2009, 10:44 PM

No, if it's wired correctly, and the jumpers are set appropriately, it will not go faster unfortunately.
Did you check the other points and stuff that has been mentioned earlier? Such as the BIOS, the hdparm tool, etc?
If you've checked everything, then there's not much more to do!

Sasha

jay73 · 07-02-2009, 11:13 PM

You can always get a PCI IDE card, they shouldn't cost too much (they are getting rare, though, so you may need to look around a bit). Or a PCI Sata controller card, many of them have both IDE and Sata (I still have a Promise Trax something something around here that has 2 sata connectors and 1 IDE connector).

Electro · 07-03-2009, 02:07 PM

Quote:

Originally Posted by GrapefruiTgirl

While this was entirely true a few years ago (not the 'stupidity' part), and is still often true for mobile HDD's and lower end or older desktop HDD's, it soon will not be true at all.

Benchmarks do not relate to real world setups. Benchmarks only show the ideal performance. Even the best hard drives such as Western Digital VelociRaptor will have the same flaws as any hard drive when ~90% of its capacity is full although this hard drive requires additional system resources. This means high performance and past hard drives still comes out the same at the end.

FYI, golgo is having problems with latency not with throughput. Performance is affected by latency more than throughput. When two different IDE drives are used on the same channel, latency gets penalized. The hard drive has an average latency of 15 to 30 millisecond and an optical drives has an average latency of 300 millisecond and higher. Since IDE devices takes turns, these drives will cause the computer to wait if the computer requires access to a drive that is paused during the transmission of data from the other device.

It will always be true even with pathetic SSD.

The following is the reason why SSD are pathetic.

1) Power problems can and will cause data corruption and/or data loss
2) Data corruption is several times higher than regular hard drives
3) Limited amount of writes.
4) The oldest data can not be any older than 10 years and this depends on quality of non-volatile memory used.
5) Data recovery is not a possibility

golgo · 07-03-2009, 04:22 PM

so thats why my computer is slowing down??well from my point of view i should really buy the adapter right or im wrong?

NeddySeagoon · 07-03-2009, 04:32 PM

golgo,

It should all work well. Setting DMA in the BIOS does not do a great deal for Linux. The kernel will do it s own thing.

Please post the output of

Code:

hdparm /dev/hda
hdparm /dev/hdb

This will tell how the kernel is controlling your IDE chipset.

As has been said, old IDE controllers run both devices at the speed of the slowest device but that has not been true for years now. Each device will operate at its own speed.

Oh, you may need to put sudo in front of those commands.

I suspect the kernel has wrongly identified your chip set and is using the generic IDE driver, which does not support DMA.

golgo · 07-03-2009, 09:24 PM

daniel@daniel-desktop:~$ hdparm /dev/hda
/dev/hda: No such file or directory
daniel@daniel-desktop:~$ hdparm /dev/hdb
/dev/hdb: No such file or directory
daniel@daniel-desktop:~$ sudo hdparm /dev/hdb
/dev/hdb: No such file or directory
daniel@daniel-desktop:~$ sudo hdparm /dev/hda
/dev/hda: No such file or directory
daniel@daniel-desktop:~

NeddySeagoon · 07-04-2009, 07:16 AM

golgo,

Thats good. It shows that your kernel is using the libata drivers and your hard drive is called /dev/sda.
SCSI does not permit DMA to be off - even when its a PATA device being driven under the kernel SCSI layer.

A few more things to check.
In your BIOS, you may have options for different modes for the disk drive controller, labels vary. IDE or Legacy is bad AHCI is good. Changing this can make it impossible to boot if you don't have the correct kernel support.

There is a kernel bug that affects some 64 bit users and not others. It involves excessive I/O wait.
Run top, press (numeral) 1 to see each core load separately and trigger the problem.

Code:

 top - 13:12:33 up 51 min,  2 users,  load average: 2.24, 2.25, 2.12
Tasks: 139 total,   1 running, 138 sleeping,   0 stopped,   0 zombie
Cpu0  :  0.7%us,  2.0%sy,  0.0%ni, 57.1%id, 40.3%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu1  :  0.6%us,  1.3%sy,  0.0%ni, 70.1%id, 28.0%wa,  0.0%hi,  0.0%si,  0.0%st
Cpu2  :  0.3%us,  1.9%sy,  0.0%ni, 97.8%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:   8176688k total,   881288k used,  7295400k free,   183264k buffers
Swap:  2088416k total,        0k used,  2088416k free,   322500k cached

look at the %wa figures.
Is from my PhenomII three core system.

golgo · 07-04-2009, 11:25 AM

top - 09:24:46 up 13 min, 2 users, load average: 0.53, 0.34, 0.25
Tasks: 166 total, 1 running, 165 sleeping, 0 stopped, 0 zombie
Cpu0 : 3.3%us, 1.7%sy, 0.0%ni, 95.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu1 : 0.0%us, 0.3%sy, 0.0%ni, 96.4%id, 3.3%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu2 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu3 : 16.2%us, 3.6%sy, 0.0%ni, 79.9%id, 0.0%wa, 0.3%hi, 0.0%si, 0.0%st
Mem: 1777024k total, 805984k used, 971040k free, 19992k buffers
Swap: 4803392k total, 0k used, 4803392k free, 308112k cached

Electro · 07-04-2009, 02:21 PM

NeddySeagoon, You are not helping golgo because golgo's setup is the following for one IDE channel.

Master: Hard drive
Slave: DVD/Optical Drive

The problem is latency. The delays is caused by latency of the optical drive during use which can be in seconds.

On any setup that I have, IDE shows its flaws when using a hard drive and an optical on the same channel. If each IDE channel is categorized as hard drive only or optical drive only, latency will not show as much.

Checking for DMA will not help because it automatically is set unless the drive is listed in the blacklist. Using SCSI for IDE makes things even worst because latency is increased during translating SCSI to IDE and IDE to SCSI. If the controller went from DMA to PIO settings and it is behind a SCSI layer, re-activating DMA is more work. AHCI will not help because it is not mature as IDE or SCSI, so performance will suffer. AHCI is not better because it is not matured enough, so I recommend using AHCI as the last resort or when there is no modules (drivers) yet for the storage controller. For a AMD Phenom or AM2+ or AM3 board, both ITE and JMicron storage controllers are supported.

The best way to fix the problem is using an SATA to IDE converter because the problem is a hardware limitation of IDE. Software can only do so much until it hits a wall.

golgo · 07-04-2009, 02:59 PM

so i should really buy the sata to ide converter if i dont want my pc to be slow????yea i have ECS BLACK SERIES A780GM-A AM2+/AM2 AMD 780G HDMI ATX AMD Motherboard, AMD Phenom 9600 Agena 2.3GHz 4 x 512KB L2 Cache 2MB L3 Cache Socket AM2+ 95W Quad-Core Processor with 2GB of ram im planing on buying another 2GB of ram would that help my pc not to be slow??

onebuck · 07-04-2009, 03:38 PM

Hi,

You need to re-read what has been stated. 'Electro' gave you some good advice relative to latency on the ide channel. Sure more memory won't hurt but you are still going to have a problem with just having a single channel ide. That's one of the reason it's great to have at least two channels of ide if that is your hardware configuration. The hdd on the same channel as the optical will continue to present problems for you.

So, solution would be to move your OS off the ide0 channel. Either with a converter or adding a SATA hdd to your SATA. Then moving or coping from the ide hdd to the new SATA.

BTW, while doing this don't listen to your rock on the optical drive.