It is difficult to say how long it will take to create a capture, as not all of the time is actually used rendering the capture. For instance some time is also required to load the target website or HTML. To speed up screenshot rendering first check that you are not using a delay when generating screenshots.
Next check, what retrieval method you are using to be notified that your capture is complete? The synchronous method is much slower than the asynchronous method as it has to poll the GrabzIt every three seconds to determine if a screenshot is complete while the asynchronous method will call your callback URL as soon as the screenshot is ready.
Other techniques that can improve capture time include:
- Blocking adverts, this reduces the time it takes to capture some web pages by 50%.
- Use GrabzIt's CDN rather than downloading the image to your server and then displaying it.
- If you are trying to capture a website that is hosted in a particular part of the world specify a country near to that location.
- If you are trying to capture websites hosted all around the use GrabzIt's exclusive Geo-targeting feature.
- If you are converting HTML into image, PDF or DOCX documents try embedding any referenced images, scripts and CSS files directly into the HTML. This will reduce the number of resources that have to be loaded before the HTML can be converted.
Another issue can be the time it takes to download a screenshot. First check the size of the captures you are producing, if they are large files you could try to improve the download time by doing one or more of the following:
- Reduce the size of the image, by altering the output width and height.
- Reduce the quality of the capture.
- Use a more compressed image format such as JPG or PNG
Finally, if you control the page you are taking a screenshot of try to optimise it. This will take a little testing on your part, but could save you several seconds. Somethings to try includes compressing images, scripts and CSS or even embedding resources directly into the HTML of web page. This avoids separate requests when generating the web capture.