How would I measure the improvement in the speed of my computer that results from deleting temp files

1

In another question about the validity of the advice to delete temporary files, no answer described how they measured the speed improvement. If I delete the temp files, or allow a software program to do the job for me as part of a disk cleanup, how much faster will it make my computer run, and how would I know? Yes, I know it will vary from one computer to another based on many many factors, but how would I measure the speed difference?

John C Hansen

Posted 2019-03-26T18:57:00.150

Reputation: 11

What question are you referring to precisely? – Ramhound – 2019-03-26T20:25:44.570

It would only slow you down if your disk was getting full. A full disk will slow down the PC and you won't need a tool to measure it. It's noticeable. If you have a fast SSD then removing temp files will do nothing for performance. – HackSlash – 2019-03-26T22:43:29.333

Answers

2

Personally, I don't believe in that advice, partly because my TEMP folder currently contains 2270 files in 196 folders with no bad effects.

For a slow-down, you would need to run a program that:

  1. Uses the TEMP folder,
  2. Opens lots of files in it.

The TEMP folder would need to contain many thousands of files and folders for any effective slow-down. If the number of files in the folder doesn't exceed a few hundreds, then any improvement will be measured in milliseconds.

My advice : Don't bother.

harrymc

Posted 2019-03-26T18:57:00.150

Reputation: 306 093

0

I expect that the majority of those claiming a performance gain from deleting temp files have done no measurements at all and evaluations were completely subjective. Such tests are subject to the placebo effect and are of questionable value. And of those that did do measurements most would have been too simplistic to produce meaningful results.

Performance tests are often difficult and tests involving files are particularly so. The results are heavily dependent on the hardware and exactly what operations were done and the state of the system before the tests. And the tests themselves will effect the contents of the file cache and other things. And that is what makes such testing so difficult. Unless proper precautions are taken you often end up measuring something other than what you are trying to test. At the least you would need to repeat the tests several times both before and after deleting the temp files with a reboot between each test. It would likely take several hours. But I suspect the reality would be even more complex.

I would not care to devise or run such a test. And after all was done I fully expect the results would be inconclusive. Delete temp files to save disk space but don't expect performance improvements.

LMiller7

Posted 2019-03-26T18:57:00.150

Reputation: 1 849

You have stated the underlying thoughts that prompted the question. So often I read statements that we should make some change to our computer that cannot be measured. I tend not to make those changes until I find a way to validate the original claim of the hoped-for improvement. And when the disk is full, deleting the temp files is not going to solve my issues for long. When my data and programs fill my hard drive I look for an additional hard drive, archive files off-line, or rethink my need for the many programs that take up space and delete them. – John C Hansen – 2019-03-28T15:55:32.400

-2

Some applications are useful for your propose. I just listed two popular on optimizing category
1. IO bit advanced system care 12. 2. AVG tune up 2019. There are some tools to measuring your system speed by cleaning garbage files on both applications.

Reza Paidar

Posted 2019-03-26T18:57:00.150

Reputation: 262

I can't see how your post answer the question. "how would I measure the speed difference?" – Máté Juhász – 2019-03-26T20:09:32.023

Before you mark me negative. Just install one those app I mentioned. Optimizing app measuring maintenance and speed of your system by percentage gauge. – Reza Paidar – 2019-03-27T21:17:28.207

It isn't that hard to devise some sort of test to evaluate performance improvements. But does the test have any real value in the real world? I strongly suspect not. I would not install anything from IO Bit for any reason. – LMiller7 – 2019-03-28T16:50:28.453

Neither one of these programs measures the speed of the computer or any software such as a browser. I am sorry, but you did not answer the question about how to measure the performance of the computer. It would be like training for a marathon race and not using a stop watch while you are training. You would never know how fast you ran each of the training runs in order to compared it to the last one that you ran. I will not down vote you but I am asking you to re-read the question and learn to read the questions before you try to promote a product. – John C Hansen – 2019-03-28T16:59:38.873