Peformance Question
Hi all,
I have some views and I have to create those views using not less then 10 tables. As each column is a lookup column. For each column I have a query Following is an example with one column. Can anybody guide me how I can increase the performance using query as column or joining with the table(s) in from and putting a outer join query.
1. select
(select desc
from lookuptable
where lookupid = b.lookupid) as column1
from main_table b;
2. select a.desc
from lookuptable a, main_table b
where a.lookupid = b.lookupid;
I would like to know which of the case will give better performance.
Thanks in advance
Ghulam Mustafa Butt
If you take as an example a case where the lookup table has very few rows and the main table has relatively many rows, then for the second query the optimizer might choose to use a hash join instead of a nested loop join, if it would be more efficient to do so. In the first query I believe that a nested loop will always be performed.
Similar Messages
-
Peformance question TKPROF/Autotrace Cost High when using index why?
I'm going through a legacy database application to do some application code performance tuning of the PL/SQL.
I am constrained from changing the code logic so I can only add indexes and hints. As I identify full-table-scans I'm using hints to force the code to use the index and now the Cost is Higher than with full table scans. Though the application code runs in batch mode and processes a lot of records.
Can anyone provide any insight into why the cost goes up when using Hints in the SQL?
I have additional testing to do though I just wanted to pose the question.Firstly, the cost can be less for a full table scan because with a full table scan Oracle can do multi block IO with a full scan. So reading a large amount of data from a table will go much faster if the table is full scanned as opposed to accessed by an index (your forcing the query with a hint can actually be extremely detrimental to the query).
Secondly, if you are not allowed to make any code changes, i'd recommend not taking the path you are currently on (adding hints). That may be warranted in a few situations, but likely not many assuming you have accurate statistics set up on your database.
Here's an article that deals with full table scans ....
http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:9422487749968
Hope that Helps. -
IF Statement Peformance Question
I have an abap program that performs 20% better with a simple change in the structure of the if statement. Anyone got any explaination why this would be so? I would really appreciate any insight into this.
This code:
IF sy-tcode IN r_order_tcodes.
...some logic #1....
ELSEIF sy-batch EQ 'X' OR
( sy-cprog = 'ZSR_ECAT_GENERATE_V3' ).
...some logic #1....
ENDIF.
executes 20% faster than this code:
IF sy-tcode IN r_order_tcodes OR
sy-tcode IN r_ecat_tcodes OR
( sy-batch EQ 'X' AND
( sy-cprog EQ 'ZSR_GENERATE_ECAT_V3' OR
sy-cprog EQ 'ZSR_ECAT_SINGLE_CONFIG'
...some logic #1...
endif.
Thanks.
ScottHi,
I believe the tcodes in r_order_tcodes and r_ecat_tcodes will have the same programs given in the else condition...
Looks like This following simple IF statement will work for all the scenarios..
Try this..
IF SY-CPROG = 'ZSR_GENERATE_ECAT_V3' OR
SY-CPROG = 'ZSR_ECAT_SINGLE_CONFIG'.
...some logic #1....
ENDIF.
Thanks,
Naren -
RH6 Extremely Poor Peformance with Lage Webhelp Project
We recently upgraded to RH6, which has caused performance
problems so severe that we literally cannot work with the project.
Our problems are exclusively with our largest project -- 600+ html
files, 500,000+ words, 35mb (only about 150 small gif images).
We're not having any problems with our smaller help projects.
When we launch the large project, the CPU usage immediately
hits 100% and stays there for the duration of the session. My
longest session is currently 35 minutes, during which I was able to
delete a folder, add another folder, and add 5 files to the system.
The application took almost 8 minutes to close.
Because of the size of this project, it's always had
performance issues on startup. Under X5, it took about five minutes
before you could work freely, but then worked fine.
So far, three of us have experience the same peformance
problems. We're all using identical Thinkpad t43 laptops with 2
gigs of memory and running XP Pro Service pack 2 with all of the
security updates. Someone recommended temporarily disabling
VirusScan, which didn't improve the situation at all.
So, two questions:
1. Have any of you experienced and been able to resolve
serious performance issues of this sort?
2. What operations does RoboHelp peform when you open a
project. It seems to verify folders (& links?), but I can't
find any information about exactly what it does.
EDIT: We are not accessing the project over a network, so
this cannot be the cause of our performance problems.
Thanks for any help you can provide.
Roger Gerbig
Senior Technical Writer
Aprimo, IncorporatedHi Roger and welcome to the RH community. To be honest your
project doesn't sound that large at all. At my last job we had
projects 4-5 times the size of that running with no performance
issues at all. That said, check out
this
link for some pointers about where you can go from here. -
Hi
I have two clients who are asking me to upgrade my computer. The type of work I do is slightly different from most people, in that I do content for video projection. Almost everything I do uses a non-standard resolution. For example the current project is 1920x3240. Currently the slowest part of the workflow is previewing and rendering. Typically the final rendering is a photo sequence of some type and encoding is done on the media server and is not part of my workflow. I used limited amounts of video footage, with most of the content created in Cinema 4D and After Effects. I use a fair amount of 3rd party effects like trapcode with high particle counts. Content is almost always broken down into small segments that are then played back through the media server, because of this, the amount of video used is pretty small.
One of my clients is willing to split the cost of a new system, so I'm willing to spend more than I normally would.
Based on reading this and other forums's here is what I'm looking at:
Case
Fractal Design Define ATX Full Tower Computer Case $129.99
CORSAIR AX1200i 1200W Digital ATX12V v2.31 $329.99
Noctua NH-D14 SE2011 140mm and 120mm SSO CPU Cooler $89.99
Fractal Design Silent Series R2 FD-FAN-SSR2-140 140mm 140mm Fan $17.99
System
ASUS P9X79-E WS $499.99
Intel Core i7-3930K Sandy Bridge-E 3.2GHz (3.8GHz Turbo) LGA 2011 130W Six-Core $560.99
G.SKILL Ripjaws Z Series 64GB (8 x 8GB) 240-Pin DDR3 SDRAM DDR3 2133 (PC3 17000) $679.99
EVGA GeForce GTX 680 FTW+ w/Backplate 4GB 256-bit GDDR5 PCI Express 3.0 x16 $539.00
Drives
areca ARC-1882I-MS PCI-Express 2.0 x8 Low Profile SATA / SAS RAID Controller Card $640.00
Western Digital WD WD5003AZEX 500GB 7200 RPM 64MB Cache SATA 6.0Gb/s 3.5" (Qnty 4) $299.99
Corsair Force Series GT CSSD-F240GBGT-BK 2.5" 240GB SATA III MLC - (Qnty 2) $479.00
areca CB-R87SA-75M Right Angle SFF-8087 MiniSAS to 4 SATA Breakout $39.98
From a drive setup I would do the following:
SSD - Boot
SSD - Cache
HD Raid - Content
HD Raid - Content
I have an external raid that has backup data.
I have a number of questions regarding the system.
Does the GTX 680 make sense or should I consider a Titan?
Will the Areca Controller make a big difference in the system for the type of work that I'm doing?
I looked at the WD VelociRaptor10K drives, would that get much of a peformance gain?
Thanks
MartinThe Areca ARC-1882i-8 has 8 internal ports, so you could go for two 4-disk raid3 or raid5 or even easier for one 8-disk raid3 or raid5, possibly with a hot-spare. Roughly, a n-disk raid3 or raid5 will give a sustained transfer rate of around 0.85 x (n - 1) x T. So a 4-disk raid3 or raid5 will give around 380 - 400 MB/s. A 8-disk raid3 or raid5 will have a transfer rate of around 900 MB/s. This shows the advantage of more disks in an array.
Software raids, whether in BIOS or in Windows are limited to striped or mirrored arrays, parity raids are way too slow. Striping is too risky and mirroring does not help speed.
Rendering for previews are helped by hardware acceleration for accelerated effects only. But the effectiveness of the video card depends to a large degree on the speed of the CPU, memory and disks. Unless the CPU is significantly overclocked to 4.5+ GHz, memory runs @ 2133 and the disk setup is very fast (400+ MB/s) a GTX 660 TI will not noticeably be slower than a GTX 680.
Rendering for export is only helped by a CUDA card when scaling, blending or blurring are applied. If you start with HD source material and export to HD resolution, there is no scaling and thus no acceleration by a CUDA card. It is all done by the CPU, memory and the disks.
If you are talking about Velociraptors 10K, no they are not faster than the HGST Ultrastars.
If you are talking about 15K SCSI disks, yes they are faster but require a SCSI controller, both of which make them much more expensive.
Despite having made the feature request for a render farm numerous times, I think since CS3, this has not yet been effected for PR. A render farm like you have in Vue would be a nice addition, but alas...
If you want to have more room for future growth, I would have a look at the 12+4 or 16+4 port ARC 1882iX controller cards. Not only do they expand on the number of ports to 12 or 16 internally, they also add 4 external ports and most importantly, they can accept up to 4 GB cache on the card, where the 8 port versions are limited to 1 GB fixed cache. Do not forget to add a BBM battery backup module to the Areca controller.
If you were to take this step and have all those ports available for future expansion, then it could be wise to have a look at the GTX 780, provided overclocking is applied. The GTX Titan is just too costly to justify. -
Raid question, can it be done or not
hi, and thanks in advance for any help given regarding my question.
i have two raptors set up as a mirrored raid array, my question is can i go into bios and change them to a stripped array (raid 0) for peformance without having to reinstall xp and if so what are the best settins?I can confirm supershanks answer, as i've TRIED it.. and no, if you change ANY of the nVraid settings, you have to reinstall windows..
I don't think this is the case with all RAID controllers though, as the promise controller on my other board, was more flexible..
you could just make a ghost image of your current setup, then change the settings, and ghost it back. check my sig for instructions using ghost with nvraid -
Want to upgrade to Mountain Lion, but CS6 questions about Auto-Save need answers...
Hi,
I'm on Mountain Lion for all my "non-critical" work, but I'm still running the latest Snow Leopard on my main desktop machine, running CS6 Design & Web Premium. I've avoided the switch primarily because of the auto-save feature that Apple placed into ML. I haven't found a satisfactory answer anywhere on the web to this basic question...is auto-save a problem, as Apple implements it, with CS6 and working on images?
In particular: 1. I work on very larger files, 4GB+ in size regularly, and I don't wan't the peformance hits associated with constant saving, 2. I make changes to images, downsizing them, croping, color corrections, etc. so that I can make jpg proofs for clients or myself. These are more "intermediary" images and not intended to be saved. I don't want to downsample an image, flatten it to creat a jpg for proof and have it overwrite the original image layered, high res by mistake. Does this happen with photoshop, or does Adobe avoid the auto-save that Apple uses?
I have all the latest iOS gadgets and it's becoming quite cumbersome managing multiple sets of iCloud info, etc. without being on ML. I'd like to make the switch, but at the cost of losing work, etc. Can someone with experience on ML & CS6 give me any input in the matter? And, does this also apply to all the other Adobe apps in CS6? I'm constantly making minor changes and edits that I usually "undo" out of and don't want saved.
Thanks.
CraigSo, in other words, photoshops behavior works the same, using save as..., etc. as it did under Snow Leopard? I should expect the auto-save behavior that occurs in apps like iPhone, etc., where i can actually lose the original because of the constant backups, etc.?
Thanks. -
Question about SAP Webgui for html
I have 2 questions regarding captioned subject.
1. I am using Webgui for html under internet explorer 10 and find that there is a red box surrounding all texts. Can I remove it ? I have no such issue in IE 9.0 and chrome.
2. I found slow peformance of webgui for hmtl, espcially the ALV grid list is displayed. It takes time to scroll from page to page.Hi,
This is SAP Business one system administration forum. Please find correct forum and repost above query to get quick assistance.
Close this thread here with helpful answer.
Thanks & Regards,
Nagarajan -
Question about SAP + SLES10 + vSphere and the truth about memory
Hi,
we are using vSphere and SLES 10 for a SolutionManager 7.0 on MaxDB 7.6.
The system is configured with 2 GB RAM in the VM. The database is configured to use 125000 pages as cache (100000 KB).
Yes, I know the system is configured very small for this usage, but it is a demo system...
But the system is really slow. I can't figure out whats the problem.
The RAM is configured with 2 GB in the VM a "free" in linux shows me that there is 500 MB used for file system chache
total used free shared buffers cached
Mem: 2076908 2002704 74204 0 28936 578212
-/+ buffers/cache: 1395556 681352
Swap: 6185016 3037744 3147272
so this shoud be ok, for me it is close to the limit...
But VMware infrastruture client shows me, that there are only 300 MB of the 2GB is used ... but the database should use 1 GB and SAP should use some RAM , too.
So who tells the truth? Where does the RAM goes to if it is not used in VMware? Could this be the performance problem?
An other problem is, that the show dev_io of the database shows me very slow access to the files ...
i.e.
I/O via Device Processes:
UNIX Devspace Read avg_read Write avg_write
tid name count time count time
28596 /sapdata/SYS/sapdata/DISKD01 6532 0.0294 101 0.2077
But VMWare shows that there is nearly no usage of DISK ... average is shown as 150 KBps...
So again, who tells the truth?
What else could be the problem?
Thank you.
Best regards
Christian> I know 2 GB is less, but the system is still 32 bit Linux, so 6 GB is "impossible".
Why that? The 32bit limit is a per process limit, not a limit for the amount of total supported memory.
> During the long weekend the system is switched to an other VMServer I am going to request more RAM. What a pitty that 32 bit does not support more than 3 GB....
I don't know who told you that but it's wrong
> But there is still the question whats correct? The linux variant of free or the VMware infrastructure client monitoring the RAM?
> What could slow down the Database IO performance?
> USE_OPEN_DIRECT we are already using ...
Are you sure it's the I/O peformance? A couple of questions arise then:
- What filesystem is underneath?
- How are your data files distributed?
- What are your I/O times for a single I/O (you can switch that on using DB50 - TAsk manager - and press on the "clock", the time will then be displayed in the database analyzer.
If you use VMWare on a recent processor you can also switch nicely to a 64bit Linux.
Markus -
Authorization question - URGENT
Hi all,
I see this in an authorization profile
Under Business Explorer - Components
infocube: *
infoarea: *
Name (ID) of a reporting component : *
Type of a reporting component : All Values
Under Administrator Workbench - Infocube
infocube subobject: Aggregate, data
infoarea: Z*
infocube: z_salescube
My question is, I have all values for reporting component under Business Explorer - Components but only one infocube under Administrator Workbench - Infocube
Does it mean that the user having this authorization has access to all the reports on each infoprovider but under administrator workbench he has only access to z_salescube infocube.
please let me know..
sabrina.Hello Sabrina!
Actually your question was not compleatly clear.
the first one Business Explorer components authorization is to restrict the user to perform certain activites on the specfied object.
i mean actually
"Under Business Explorer - Components
infocube: *
infoarea: *
Name (ID) of a reporting component : *
Type of a reporting component : All Values"
here you must see a fourth field Activity whcih specifies which activities you can perform on these Bex objects which means...
You can perfrom the specified activity on all the queries
and the second mentions.
the user can peform the specified activity on the Agregates and data of this perticular cube.
with regards
ashwin -
(not) another bootcamp question!
Hello,
Well after reading many posts and seeing the repetitive nature of them I find myself reluctant to add to the list...
However... Despite my trawling and repeat attempts keep failing dismally! So appologies in advance at asking what I am sure is a repeat question!
My Aim is to install drivers for camera and sound card for Windows 7.
I have an iMac mid 2007 running Snow Leopard 10.6.8 and Boot camp v3.04
The steps taken so far is this:
1. Ran apple software update (no results returned)
2. Opened Bootcamp and selected option to install from disc (the option to download drivers returned an error cant remember excatly)
3. Created a partition 200gb
4. Installed windows 7 on partition
Now windows seems to run perfectly with the exception of the sound issue and missing drivers for the camera, peformance counter & usb input device.
The issue I have is I no longer have the Snow Leopard disc, after reading previous posts I have downloaded multiple versions of bootcamp, 3.####, 4.#### & 5.####. When I run the setup.exe it comes back with an error saying "this version of bootcamp is not intended for this computer model"
One thing I have recently read is that in Windows there should be a boot camp icon on the start menu tray, this doesnt seem to be there and when I navigate to unistall programs it isnt there either, could this be a potential sign/issue?
So anyway, I am looking for some kind of a solution to the problem, if anybody can help?,Well, I have now managed to succeed and thought I would share the simple solution with anybody that may be having a similar issue.
This page will help identify what version of bootcamp to download:
http://support.apple.com/kb/HT5634
It will also take you to the download page so you get the correct verision in my case 4.4033, run the installer and it should install drivers for everything! -
Hi, I have hp elite m9000t series desktop with 2 GB(1 x 1GB) ram and as per the specs and research in hp website it can support to 8gb. RAM is of DDR2-SDRAM (pc2-5300 / 667 MHZ) type. I have 2 more slots open, so guess that can add 4gb(2 x 2gb) of ram. Question is 1) do i need to buy the sam ddr2 667 mhz ram or can i go to higher speed models? 2) is it compatible to have 2 slots with 1 gb each and another 2 slots with 2gb ram each? 3)what is the recomended RAM for window7 better peformance? i am planning to install VMware to do other OS like linux, etc.. given that i may need more RAM. Please suggest me on the memory upgrade..I am seriously looking towards windows7 release date, i am very much disappointed with vista, cant wait for windows 7.
btw, i am planning to upgrade to 64bit windows 7 professional, currently i am running with home premium 32bit.It's best to install RAM which is just like the one you have right now, i.e. pc2-5300 / 667 MHZ. In fact, I would recommend using RAM from the same manufacturer as the one that is currently installed. Each slot should have the same amount of RAM. So I suggest you use 1+1 or 2+2 rather than 1+2. Since you'll be using VMWare(hope your processor supports Virtualization), I suggest you go for 2+2.
Make it easier for other people to find solutions, by marking my answer with \'Accept as Solution\' if it solves your problem.
Click on the BLUE KUDOS button on the left to say "Thanks"
I am an ex-HP Employee. -
Nub question concerning threads.
I recently completed my first Comp Sci. class in Java and have decided to try to continue learning on my own. I am now trying to make a pong game, using a turorial. Unforunately my course did not cover threads in any way, shape, or form. I am able to get the beginning of the program to work (code posted below) and have no problems with it; however, I do have a few questions concerning what certain elements of the code do.
1. Can someone please give a definition of a Thread in "Dummy" speach? I've looked up definitions and I can't seem to really grasp what they are and what they are used for.
2. In my program there are two lines (32 & 79) where the thread priority is changed. Can someone explain what that means and how it affects my applet?
import java.applet.*;
import java.awt.*;
public class BallApplet extends Applet implements Runnable
//Declare variables to change posistion
int x_pos = 10;
int y_pos = 100;
int radius = 20;
boolean rWall = false,
//The second image of the ball
private Image dbImage;
private Graphics dbg;
public void init() {}
public void start()
Thread th = new Thread(this);
th.start();
public void stop() {}
public void destroy() {}
public void run ()
Thread.currentThread().setPriority(Thread.MIN_PRIORITY);
while(true)
while (rWall == false )
if (x_pos == 700)
rWall = true;
else
x_pos= x_pos + 5;
// repaint the applet
repaint();
try
// Stop thread for 20 milliseconds
Thread.sleep (20);
catch (InterruptedException ex)
// do nothing
// set ThreadPriority to maximum value
Thread.currentThread().setPriority(Thread.MAX_PRIORITY);
while (rWall == true)
if (x_pos == 0)
rWall = false;
lWall = true;
x_pos = x_pos - 5;
repaint();
try
Thread.sleep(20);
catch (InterruptedException ex)
Thread.currentThread().setPriority(Thread.MAX_PRIORITY);
public void update(Graphics g)
//initialize buffer
if (dbImage == null)
dbImage = createImage(this.getSize().width,this.getSize().height);
dbg = dbImage.getGraphics();
//clear screen in background
dbg.setColor (getBackground());
dbg.fillRect (0, 0,this.getSize().width,this.getSize().height);
//draw elements in background
dbg.setColor (getForeground());
paint(dbg);
//draw image on screen
g.drawImage(dbImage,0,0,this);
public void paint (Graphics g)
g.setColor (Color.red);
g.fillOval(x_pos - radius,y_pos - radius,2*radius,2*radius);
}Thanks for any help, it is greatly appreciated.Questie,
Please control yourself. Please.??
As the program runs it is one thread. (You canthink
of each thread as an independent program if youlike.
This is not 100% accurate but close enough that
it
might help you understand)
It will only confuse the OP
You are a flat out wrong for a number of reasons.
1) You seemed to miss the part where I did say that a
doubling in peformance is fictious and is dependent
on a number of factors.As i was telling that would only confuse the OP.Its better to be 100% precise when you explain to an OP
2) You don't understand how a multi-process OS works
in the least.Assumptions as usual
3) You don't understand how processors work in the
least.Well,what was i saying?
2 and 3 are important because your point about
context switching is not valid. First of all because
this happens anyway because your program is not the
only thing running the processor. Second because of
the first point each thread gets a timeslice, you may
with multiple threads each get a timeslice and these
slices may (and often do) add up to more total time
then the first got by itself.When you consider a solution to a problem the average and the worst case has more weightage than the best case.You are taking the best case assumption here-"because ofthe first point each thread
gets a timeslice, you may with multiple threads each
get a timeslice and these slices may (and often do)
add up to more total time then the first got by itself".Now dont stop whining that you werent.Never take the best case always coz its the average and worst case which checks if the system festers to a condition beyond rectification
For example (these numbers are fictional but I am
trying to explain it so don't go off all half-cocked
again please)
Program with one thread. Each thread gets 100 cycles
per second.
Program with four threads. Each thread gets 50 cycles
per second.
The second program will finish in half the time as
the first. Because even though each thread is slower
as a whole the program is getting twice the cycles
of the processor as before. You are taking the best case again and moreover its the cpu scheduling algo which decides the time slice and not you sitting on the CPU.I have a vague feeling that you have an idea that you are sitting on the CPU deciding the Time Slice.
You need to understand something important here. Alot
of the time when you are doing things in a program,
and this doesn't just apply to IO here, you don't
have to use the processor for each step. There is
some shifting around of things to get them ready to
be executed by the processor. A more advanced
representation of this is to be found with
Hyperthreading. Hyperthreading processors (single
core) are of course single core but they act as if
they were multi-core (multi processes) because while
only one process is actually executing at one time
the others are pipelined in such a way to make sure
that the executing core of the processor has the
least amount of idle time possible.Digression
So that means on a basic processor each time you add
2 numbers it takes 5 steps. And the processor core is
idle 80% of the time.
From where did you get that 80%?Stop coming up with your own numberes.This is no number game
So for example while some data is being placed in the
outbox new data is also being collected and is at
step 2 already. In this model while we are still on a
single core the core is now idle only 60% of the time[b]From where did you get 60% now
So yes threading is very useful for IO and other
operations that may block your program but they have
more use than just that. Parallel execution can and
will have some performance advantages (if designed
properly) even on a single core processor without
hyperthreading. In a well written multithreaded
program when you add hyperthreading and multiple
cores or multiple processors to the mix then you
really have something.I never said it wont.Read properly what i have posted.I told dont take the best case always.You need to consider that too but the other two carries more weightage
Please don't feel you ever have to post to correct me
again thanks.^^
Get ride of this self conceited attitude.Even if you have it dont make it public.Its not that you are always right cotton.I am not saying that you are wrong.If you have read my first reply i mentioned that i knew that you gave that example just to make the OP understand.The problem with that is that if the OP is unable to fathom out what you mean he is going to misconstrue whatever you have said and that propogates -
Set-up As follows:
AMD Athlon 1000Mhz Socket A CPU with Coolermaster Fan
512MB DDR333 PC2700 RAM
MSI KT4 Ultra - SR Motherboard with optional S & D Brackets
Liteon 52246S CD-RW (52x24x52)
Samsung 12x DVD-ROM
MSi Nvidia GeForce4 Ti 4200 GFX Card (AGP 4x, 41.09 Drivers)
Seagate Barracuda 40Gb as master on IDE1
Maxtor Diamond Max 7200rpm 60Gb Drive as slave on IDE1
Creative SoundBlaster 5.1 Digital PCI
Creative DTT2200 5.1 Surround Speakers
Enermax CS-305 Black Case with Side window
Windows Xp Proffessional (no SP1)
A 350w Powerup powersupply (one that comes with the enermax cs-305 case)
Several Questions:
Processors/memeory
I thinking of changing the processor to an athlon xp. With one should i get? And why is that those processors which have the 333FSB are nearly £90 more expensive??? Does matching the fsb speed to your memeory speed increase peformance too?
Powersupplies
Is my current powersupply sufficient for my setup?
Is this PSW any good_ Q-Tec 550w Dual Fan (that is quite cheap)
And why is this one much more expensive? Enermax 460w
Cooling
using Fuzzy logic 4, the temp displayed is mostly 55-64. Is this too high? And can someone please explian how i find out the full load temps etc.
GFX Cards.
I have the 128MB agp 4x MSI GeForce4 ti4200 card running at 250 core/459 clock usign the MSIO clock tab in desktop properties. Can this safely go any higher?
USB Ports.
On my M/B i can get the top two ports to work fine, but not the bottom two. How come. I have the bracket installed also with the extra 2 usb ports - does this have something to do with it?
Thanks in advance.1. KT4 Ultra can support upto the fastest XP2700+, you can get it provided you have money... Yes, FSB matching with memory speed the fastest.
2. Enermax has extra features such as fan-control adjust.
3. If it's idle, it's high. to get full load, just run some CPU benchmark and loop it.
4. Your question sounds like "if I buy lottery today will I win?" No comment on the overclockability.
5. Have you enabled ALL USB PORTs in the BIOS? -
N570GTX Twin Frozr III Power Edition Questions
Hi,
I just have two questions regarding my card.
Firstly, is this http://www.megafileupload.com/en/file/427489/MSIN570BIOS-bin.html the most up to date vbios for my card, if not where can I get the latest one from?
Secondly and this is probably a stupid question, my card has a switch on it that switches the cards fan mode between Performance and Silence. Which is which on the card? I have tried the switch both way but I couldn't really notice a difference?! If I was looking down on it with the backplate to my left, which side of the switch would be on the left?about second question, they are named [peformance & silence]
see it on image:
http://media.msi.com/main.php?g2_itemId=65220&g2_imageViewsIndex=1
when you switching between BIOSes, you need to reboot your computer,
then the new BIOS will be active.
if you switching them without reboot, they are not switched.
about 1st one
update is available, i've attached it below.
Maybe you are looking for
-
Can I use Facetime between my MBA and my ipad?
-
How to open an .indd file in commands
does any one know how to open an .indd file in commands from a plugin(without opening from the normal procedure..File -> Open ...) in my plugin I tried, SDKFileOpenChooser fileChooser; fileChooser.AddAllFiles(); fileChooser.ShowDialog(); It sh
-
There are too mant TREX Errors
Hello Experts, I have created an Index for a specific folder in KM. Under KM--> Sys Conf --> KM --> Index Administration it shows the green led, which I believe that the data sources are completely indexed. But, under the KM--> Monitoring > KM> Index
-
I cannot find /Library/internet plug-ins as mentioned in a previous forum
-
Problem downloading additional softwares
Hi everybody, When I try to download SAP Add-in for Crystal Reports 2011 the following message appears: Same problem with any other additional software or user. Does anyone know why? Thanks!