Slower search response question

[Martin H]

Simon - as Anton says - it does overwrite but equally his proposal should also work.

Bit confused by the differences then. I will do a structured, measured test and see what shows up. I have done the testing 'informally' to date but would be confident it is directionally correct because it has been so consistent. I did a very short comparison between the mini with live vs the air with the beta and it was 3 or 4 times as slow side by side at best.

[antonmb]

I'm not sure if it would make a difference, but for what it's worth all my devices, including the iPad Air have assigned ip addresses.

[Martin H]

Anton - I can try that. The servers and PC IPs are all tied to their MAC addresses and I can add the ipad air if you think its worth a try. Who knows. It is odd.

[Peter@57m]

Martin,

I have 10,000 tracks in 720 albums and search on my iPad Air iOS 9.2 running Kazoo beta is one second or so. MinimServer is running on a Melco N1A.

KDSM, Melco, iPad etc all have fixed IP addresses

Peter

[Martin H]

Peter that's the same setup as me except for the fixed IP on the Air so I will definitely give that a go.

For info - I currently have just under 3700 albums and 48000 tracks (which sounds crazy now I write it down). So I could expect it to be a little slower but most of the time before it was only a second or two at most really on every device including the PC. Often displaying partial results before I could count. Maybe slightly slower on the idevices but nothing to worry about.

With the beta the idevices appear to be handling the search markedly slower than the PC now. Hence my confusion. If I had to guess now I would have to assume that the beta has either changed the way the query response are made and processed, that the level of underlying memory usage is greater and is impacting this processing or there is some underlying increase in processing power on the IOS beta that's impacting it. The first 2 sound more likely than the latter.

The minimal side by side testing of the live to beta was equally distorted with the beat slower (mini vs Air beta) and if anything I would expect the Air to out process the mini (original).

Unfortunately I don't think I will be able to complete a better documented test until tomorrow or the day after unfortunately but it would be difficult to see why this would be different answers from the normal usage and random tests I have done to date.

Thanks for the support everyone