ASRock CoreHT 252B Review

The CoreHT 252B represents a quantum leap in both CPU and HTPC related GPU prowess over the Core 100. As such, we have no hesitation in recommending the CoreHT if the Core 100 was ever in your radar. Of course, pricing is an important aspect. We are assuming that this mid-range HTPC will debut around the same price with which the Core 100 was launched.

Our recommendation, however, doesn't come without reservations. Both ASRock and Intel have areas in which they can improve. Throughout the review, we have covered the positive aspects of the CoreHT 252B.
In this concluding section, we will identify aspects which can be improved.

ASRock:

1. The performance of the HTPC could have been improved by using a SSD for the boot drive. ASRock indicates that a second 2.5" hard drive can be attached to the underside of the ODD/HDD chassis. It should be pretty straightforward to use a 32 GB SSD as the second drive without having an exorbitant rise in the unit's price.
2. The supplied MCE remote is basically a marketing checklist item. Instead of that, a mini wireless keyboard / touchpad combo would have also made up for the lack of a wireless keyboard and mouse in the package.
3. Minor industrial design updates to make the chassis more attractive would be nice. For example, the ASRock logo could be shifted to the lid and even made bigger in the process. Similarly, the power LED can also be shifted to the lid (similar to what is in the Zino HD 410)

Intel:

1. Intel's DXVA implementation is a very big mess for all the open source DXVA decoders (MPC Video Decoder and XBMC's DXVA2 implementation). Intel should step in and work with the developers to make sure those decoders work perfectly with the Intel GPUs. Intel GPU based system purchasers are forced to shell out more money for commercial decoders in order to get the benefit of reliable hardware acceleration even for local file playback.
2. It is not possible to get hardware acceleration for decode as well as madVR rendering with Intel GPUs and open source / freeware decoders. Intel's GPU shaders are also not powerful enough for consistent madVR. We hope that the GPU in Ivy Bridge will be powerful enough for madVR so that we can contemplate running some benchmarks for that.
3. Needless to say, the video post processing algorithms can use some improvements to get to the quality provided by the AMD and NVIDIA GPUs.
4. Intel needs to make the custom resolution feature user friendly (at least as good as NVIDIA's). It must not be restricted by the EDID from the display (for the advanced users, at least)

Despite the above issues, we were able to configure the HTPC to serve the average user well. As that is the intended goal of this mid-range HTPC, we have to say that ASRock has succeeded in putting out a capable successor to the Core 100.