Performance issue in downloading the images on iPhone from the server

We have an application on iPhone. This application displays 25 products per page/screen. The text items such as product name, price, discount, URL of the product image, etc of all the 25 products is downloaded from the server first.

After that we make 25 synchronous requests to download the 25 product images, one after the another. Each image is about 25KB in size and these are of size 300 by 400 pixels approximately and we only need 72 by 72 pixels size images for display on iPhone. We notice that it takes about 40 seconds to display one screen/page and this sort of performance is not good. So we are investigating how to increase the performance.

1. Will the performance improve if we scale down the size of the images on the server to 72 by 72 pxiels.
2. Also is it possible to download all 25 images from server to the iPhone? If so can you please share your approaches as to开发者_如何学编程 how to do that? We want to do this only if it can improve the performance.

I suggest you ask this on Stackoverflow as already mentioned.

From a programmer's perspective, if you only need 72x72 images, you should definitely bring that. You'll be saving bandwidth, battery and processing power.

Then 25 sync requests seems like a bad idea, why not bring an entire page (or two pages) at the same time?

A URL request is "slow" by nature, so the less you make, the faster it will work.

I'd modify the server to allow batch fetching as in "give me the first 25" and then you process them locally. Then you can fetch the next 25 asynchronously (and preemptively) for when the user presses next you already have it (and always have 1 or 2 in advance).

Use Cache, save them locally if you can and always check if that page is available locally, so you don't have to re-fetch the records if the user presses back and then next again. What's downloaded stays downloaded :) Product pages should't change that often.

For more specific implementations, I suggest you jump over to StackOverflow.