Write Cache Enabled.

MadOblivion

MadOblivion

Weaksauce
Joined
Jan 24, 2011
Messages
108
Just enabled write cashe on a two drive raid 0<disabled is default>, here are the results on crystal disk mark. Looks like the Write speed was improved significantly and the majority of the read speeds as well except sequential read.

Enabled-
WriteCacheEnabled.jpg



Disabled
ICH10Rbench.jpg
 
Last edited:
fXK said:
Still slow my raid 5 is 600 write and 600 read
Click to expand...

I dont run SSD's bro and if you don't run ssd how many drives are in your raid 5? It cant be only two drives... i would have to call bs

Why don't you troll somewhere else? you compare your raid to mine, yet you run a different raid all together with hardware you fail to mention.
This thread is really just made to show the difference in performance with write-cache setting.. So popping on here and saying your drives are faster and giving no viable technical information that even pertains to this thread is just ignorant.

I
 
Last edited:
Dont believe me? here it is, Nope not SSD either. Plus for raid 0 and 50mb test speed LOL that speed is bad. My raid 5 probably pull1000mb+

1sgmc4.jpg
 
report this guy for being a retard. Again you fail to mention how many drives you run and what type.

If you have no useful viable information besides a benchmark. You obviously run more then two drives, 8 or more? and if you know anything about raid your read and write increases with the number of drives you add to your array.

I feel my IQ dropped a few points just reading your posts.
 
Last edited:
fXK said:
Dont believe me? here it is, Nope not SSD either. Plus for raid 0 and 50mb test speed LOL that speed is bad. My raid 5 probably pull1000mb+
Click to expand...

The point was for 2 drives. I could post 900 MB/s reads from my 9 drive linux software raid 6 at work but that would not mean anything in the discussion.
 
If software raid can pull 900mb/s what the point of expensive hardware raid?
Click to expand...

For users running on windows where a hardware raid card is needed for good performance because windows software raid does not effectively cache the raid stripes. Without this cache writes suck. On non windows operating systems the cache is effective but opens a chance of data loss if power is lost when the cache has dirty data.
 
MadOblivion said:
Just enabled write cashe on a two drive raid 0<disabled is default>, here are the results on crystal disk mark. Looks like the Write speed was improved significantly and the majority of the read speeds as well except sequential read.

Enabled-
WriteCacheEnabled.jpg



Disabled
ICH10Rbench.jpg
Click to expand...

How about benchmarks using greater than 50MB.
 
Just a reminder in case anyone forgot

(1) Absolutely NO FLAMING, NAME CALLING OR PERSONAL ATTACKS. Mutual respect and civilized conversation is the required norm.
 
For my next question,

What espect of crystaldisk mark will offer my more performance under normal use for gaming ,apps and video encoding, The 50 mb test or the 1000mb?

I do notice a small dip in performance with write cache enabled on the 1000mb test in both read and write. As for the 50 mb test the Write speed increases substantially..

Will write cache slow down or speed up video encoding? I really dont have a great deal of understanding when it comes to this.

I do know at any given moment most files being read or written are rarely above 50mb. Example, Battlefield.exe is 32mb

p.s. thanks for the reminder Crosshair and Thread hijacker control
 
I have no time to fully answer any of your questions. I need to get to the day job.. Anyways 50MB can show unrealistic benchmark numbers because of cache effects on the drive and the OS.
 
drescherjm said:
I have no time to fully answer any of your questions. I need to get to the day job.. Anyways 50MB can show unrealistic benchmark numbers because of cache effects on the drive and the OS.
Click to expand...

Well i run 50 mb tests for a couple reasons. One being 90% of the files on a computer are usually under 50 mb. The second reason is that most subsystem HD testing performed by HARDforums review in crystaldiskmark is 50mb or 100mb.

I was looking to compare my Raid performance with a system with same specs and HARD reviewed a Asus Rampage III formula with drives and cpu model identical to my own.

I'm assuming under normal use the 50mb/100mb test should be more then adequate, but i have a feeling video encoding might reflect more so on the 1000mb test. I don't know how the encoding process works to know for sure, Do video encoders write data in parts and combines the data after encoding is complete? Or are the files encodied in one big gb shot?

I suppose i can use a stop watch to compare encoding times.
 
Last edited:
MadOblivion said:
I was looking to compare my Raid performance with a system with same specs and HARD reviewed a Asus Rampage III formula with drives and cpu model identical to my own.
Click to expand...

You wanted this but didn't list any of your specs, drives or hardware at all. Is English not your primary language? This may help in trying to figure out what you are asking...
 
^ I was looking for a spec posting, but didn't see anything but an affiliate link in the sig.

MadOblivion, that affiliate code might be against the rules:
(19) ADVERTISING, site pimping, contests or any type of business promotion is not permitted. Soliciting for fraternal organizations, humanitarian causes or personal rewards is prohibited unless approved prior to posting.
Click to expand...
 
ob1 said:
You wanted this but didn't list any of your specs, drives or hardware at all. Is English not your primary language? This may help in trying to figure out what you are asking...
Click to expand...

What do you think i was asking? If you read the article my questions are about crystaldiskmark. My system specs should not be of any importance to the questions posed..

If you would read how the sentence in the qoute is worded, it is clearly past tense and not posed as a question

"I was" looking to compare my Raid performance with a system with same specs and HARD "reviewed" a Asus Rampage III formula with drives and cpu model identical to my own.
example:" I Was" not "I Am"
example:"reviewed" not "review"
 
Last edited:
The reason you aren't getting the answers you want is because of a few things, one being your post composition and most people don't understand what you want to know. The second thing is the way you are coming off in this thread. I have a good grasp of the langauge and I don't need lessons, but your post is hard to figure out unless english is a secondary langauge to you. I was trying to establish this to help you out, not get an english lesson.

I wish you good luck and hope you find the answers that you seek...
 
ob1 said:
The reason you aren't getting the answers you want is because of a few things, one being your post composition and most people don't understand what you want to know. The second thing is the way you are coming off in this thread. I have a good grasp of the langauge and I don't need lessons, but your post is hard to figure out unless english is a secondary langauge to you. I was trying to establish this to help you out, not get an english lesson.

I wish you good luck and hope you find the answers that you seek...
Click to expand...

Thank you for letting me know i need a English lesson. I was merely giving you a reading comprehension lesson, works both ways doesn't it?
I think you should probably stick to the topic at hand, you are analyzing a post that was not meant for you to reply to. Do you think that is why you have trouble understanding the post?
I didn't create this thread to give system specs or to debate benchmarks. The subject is very simply What is the benefit and cons of Write-cache, just like the Thread topic reads, discussing anything else is not needed nor requred . That is all, have a nice day.
 
Last edited:
MadOblivion said:
Well i run 50 mb tests for a couple reasons. One being 90% of the files on a computer are usually under 50 mb. The second reason is that most subsystem HD testing performed by HARDforums review in crystaldiskmark is 50mb or 100mb.

I was looking to compare my Raid performance with a system with same specs and HARD reviewed a Asus Rampage III formula with drives and cpu model identical to my own.

I'm assuming under normal use the 50mb/100mb test should be more then adequate, but i have a feeling video encoding might reflect more so on the 1000mb test. I don't know how the encoding process works to know for sure, Do video encoders write data in parts and combines the data after encoding is complete? Or are the files encodied in one big gb shot?

I suppose i can use a stop watch to compare encoding times.
Click to expand...

I don't need a reading comprehension lesson, because I know what a question mark looks like, and you put a few in your post, so I had assumed you had a question. Its obvious now that you are trying to troll people here, not discuss anything relevant to the topic. And you can keep editing your posts all you like, they are still hard to understand as either your spelling skill and/or reading comprehension skill needs work. We can do this all day until we both get banned...
 
ob1 said:
Its obvious now that you are trying to troll people here, not discuss anything relevant to the topic.
Click to expand...


Interesting you accuse me of trolling, yet I don't believe you have mentioned anything about write-cache, that is the thread topic. You bypassed the first post and try to involve yourself with a post that was furthest from the topic at hand. Correct me if i am wrong but is that not the definition of trolling.

And yes you should be banned right along with Fxk
 
I will stick my oar in here only to the extent that I find it irritating that "troll" seems to be one of the first words people resort to when they don't like what/how someone is saying something. Do we really need to get into a flame war here?
 
danswartz said:
I will stick my oar in here only to the extent that I find it irritating that "troll" seems to be one of the first words people resort to when they don't like what/how someone is saying something. Do we really need to get into a flame war here?
Click to expand...

I know, i feel like I somehow lower myself to his level by even replying. He obviously has no constructive input on the thread topic.
 
I was trying to help you, wow. I figured by your spelling and grammar you might have been foreign, so I was trying to figure out exactly what you were asking. Again, I hope you find what you are looking for and get the answers you want...wow
 
You must log in or register to reply here.
Back
Top