Skip to content

One Week LPM Roller Coaster Ride: From 5 to 100 back to 4

jerryjijerryji
edited March 2013 in Bugs
v5.35 is as stable as rock, but it seems to under-utilize my 200 threads and churns out only 4 LPM --

image

While there has been no change to my settings that used to give me 100+ LPM (details: https://forum.gsa-online.de/discussion/2611/my-gsa-ser-config-full-monty-warning-low-lpm#Item_16 )

@sven I see quite a few others complaining about low LPM with the new versions, if you need my log or something to debug, please shoot me a message here. Thanks.
Tagged:

Comments

  • lrichard2112lrichard2112
    edited March 2013
    image

    mine too was not good.

    i want to roll back my SER version but I dont have the serial for SER yet :(

    @sven how to make GSA send me again the serial of my SER?

    tnx
  • SvenSven www.GSA-Online.de
    You know what? I will remove this LpM shit, people rely to much on this numbers instead of watching the actual results.
  • jerryjijerryji
    @sven , I'm cool with removing LpM. Just wish you can clarify what this "actual results" you were referring to -- I sure hope it's not SERP because it's just too far away.

    BTW, if you remove LpM, you might want to remove the thread count as well because (other) people will complain if the thread count always falls way below their setting.


  • eLeSlasheLeSlash
    edited March 2013
    Of course he wont remove thread count. Also @sven implemented the LPM just to make a basic ideea how many links is gsa ser submitting(i think). The actual results comes from your verified links ammount. You can reach 100 lpm easily if you only use blog comments for example. I never pass 25-30 LPM and i create around 6-9.000 verified links in 3 hours mostly are incontent backlinks (not blog comments).
  • OzzOzz
    edited March 2013
    i don't mind if it will be gone either. most people don't realize how to assess the numbers and start freaking out when other have higher numbers without knowing what kind of settings other people are using.
  • AlexRAlexR Cape Town
    You can remove the LPM count. It's a random number in my eyes. Who cares about LPM? 

    About thread count - I'd much prefer an "average thread count" number. This would add more value since depending on what SER is doing, the thread count changes. At least with an average I can actually see what it's using over a 24 hour period (i.e. it covers searching, submitting, verifying). 

    With this "Actual Thread Count" - the number depends on when you view it! 
  • jiggsawjiggsaw
    edited March 2013
    I don't get all the "remove LPM" comments. When I had a higher thread count/LPM my submission numbers were higher. Now that I have a lower thread count/LPM, my submission numbers are lower. So even if LPM is gone, my "actual results" are still lower than they used to be. So in my opinion it was helping me gauge my progress. Unless I'm crazy and the submission count doesn't mean anything either.  :-/
  • Corleone72091Corleone72091
    I agree with Jiggsaw. I ran 5.34-5.36 with lower submissions and lower verified numbers in addition to the lower Lpm and lower threads. So as far as actual performance goes, my numbers were way lower with the last few updates.

    Downgrading to a copy of 5.26, my submission and verified numbers are more than 3 times higher as compared to the newer versions. Lpm seems to have a high correlation with submission and verified numbers so I would like to see it stay since "actual results" seem to decline with Lpm.

    What I would like to see is a verification success rate.
  • OzzOzz
    edited March 2013
    the thing is not about LpM counter itself but how inexperienced users interpret it. when SER runs in its cycles the number vary and inexperienced users freaking out if they notice when it is below their "all time high" number after a couple of hours.

    there is the submission counter as well that gives you an idea how things are developing as well. apart from that an average submission counter of the past 24h might be more useful like AlexR said. those short term results/numbers will get you nowhere in my opinion.
  • eLeSlasheLeSlash
    Why dont you guys understand that if some times passes and a specific project was active long time and posted to quite alot of websites, it will be more and more harder to find new targets to post to, of course your submissions will decrease as long as the verified number goes up. Atlest thats what im experienceing.
    @Jiggsaw that doesnt tell us nothing. Maybe you had different projects active with different platforms selected, maybe your proxies were other faster, maybe you used imported target lists, maybe you used global lists... things differ you cant just compare that easily
  • jiggsawjiggsaw
    edited March 2013
    Please... I'm not a new user. I've been using GSA since it was a NoHandsSEO rival.  I know what I'm doing. The fact that I rolled back to 5.26 and changed absolutely nothing means something is different. Same active projects, same private proxies, same threads, same captcha software, etc. LeeG is also a veteran user like myself and he's experiencing the same thing. So yes, I can just compare that easily. 
  • ComputerEngineerComputerEngineer
    instead of lpm processed url count would be a lot more accurate
  • DavidA2DavidA2
    While I agree that there is (sometimes) too much emphasis put on the LpM figure, and I understand the frustration that @Sven has has with having people complain about it, I don't think removing it is the answer.

    LpM is (when used properly) a reflection of the throughput that SER is performing. The fact that users can switch between releases without changing anything else and see such dramatic differences in these figures seems to indicate that there is a snag somewhere with the newer releases. To deny this doesn't seem logical.

    I have been sticking with release 5.28 because of the problems that others have been having. This release seems to be maxing out my CPU, but I still get good performance.

    I will probably stay with this until I see that things have "settled down" somewhat.
  • jiggsawjiggsaw
    Thanks @DavidA2 That's all I was trying to say.
  • CecilDeeCecilDee
    edited March 2013
    Hey @jiggsaw & @DavidA2, what is the process involved in going/rolling back to a previous version of GSA?  How exactly is that done?
  • sonic81sonic81
    just download the binary into the gsa program folder and run it.
  • jerryjijerryji
    Where can I download previous version binaries?
  • DavidA2DavidA2
    When you do an update of SER, the version that you are currently running is saved (renamed) in the folder that you are running SER from (probably "C:\Program Files (x86)\GSA Search Engine Ranker\"). It is renamed "Search_Engine_Ranker_old.exe". You can run this version of the software anytime just by clicking on it.

    To get any other versions, you can search the forum. I think others have posted links to some of them here.

    Hint: If you want to have the option to go back to the version you are currently running, make a copy of "Search_Engine_Ranker_old.exe" and rename it to "Search_Engine_Ranker_x.xx.exe" (where x.xx = the version number). But...you must do this BEFORE YOU UPDATE THE PROGRAM or this file will be replaced with the new version. (Note: If you forget, you can always rename "Search_Engine_Ranker_old.exe" to "Search_Engine_Ranker_x.xx.exe" as well.)
  • CecilDeeCecilDee
    edited March 2013
    @DavidA2 Thanks, good info to know.  Another thing, seems like the search doesn't find numbers?  If 5.22 or any number for that matter is inputted into the search bar, there aren't any results.  How do you guys search when numbers are involved?
Sign In or Register to comment.