If you are on this page then you are either lost, inquisitive or want to learn about delta temp graphs. Well if you are the last one, then you are on the right page, if you are one of the first two, either hit your browsers back button or stick around for a read.

Let’s start from the beginning (Always a good place to start). You will have no doubt seen that our new temperature graphs seem to be showing very low results. This is not because there have been major cooling breakthroughs recently, but it is simply because we have switched to a fairer way to show CPU/GPU cooler results.

Here is an example delta graph:

Delta is a Greek letter and looks like this: Ä . In science and maths it is used to represent a change in something. So here it is used to mean change in temperature. Change in temperature against what you ask? For all of our graphs it is change in temperature against ambient (room temperature).

To interpret the graph, you read it as follows:
• The stock heatsink kept the processor at 12oC above ambient at idle, and 34oC above ambient at load.
• The Akasa AK975CU kept the processor at 9oC above ambient at idle, and 25oC above ambient at load.
• And so on…

As with normal CPU/GPU cooler graphs; the lower the numbers, the better the cooler performed.

The next question you probably are asking is why we need to use these instead of normal CPU/GPU cooler temperature graphs. To answer this requires a slight bit of explanation.

Here is what the normal CPU temperature graph would look like from the same results above.

These results were recorded over a long period, and as a result the ambient temperature varied by a couple of degrees (we don’t have fancy labs to test our hardware in with climate controlled rooms!). Now because of this temperature change it has skewed the results and made it look like certain coolers have performed better/worse than they actually did.

Let me now show you how the temperatures would look if the ambient room temperature had been kept at 25oC throughout the test.

If you compare the two graphs you can clearly see that they are not the same. While it is only a degree or two in this example, and doesn’t actually change the position of the coolers in performance terms, it could be a lot more in different tests. Say for example the first cooler might be tested on a mild day where ambient temperature is 21oC, but the second cooler might be tested on a hot day where the ambient temperature is 25oC.

It is unfair to compare the performance of a cooler with ambient temperature of 21oC against the performance of a different cooler with ambient temperature of 25oC. This is because the first cooler gets a 4oC advantage over the second one, and in graphs, this difference in ambient temperature is not shown. So it is for this reason that we are using delta graphs to show the performance of the coolers.

I hope you like the new way if not drop a line to reviews (AT) xtremecomputing.co.uk