HEXUS.bang4buck, temps, high-def decode, overclocking
HEXUS.bang4buck
In a rough-and-ready assessment of the cards' bang per buck, we've aggregated the 1,024x768 frame-rates for the three games, normalised them* and taken account of listed the cards' prices.
But there are more provisos than we'd care to shake a stick at. We could have chosen three different games, the cards' prices could have been derived from other sources and pricing tends to fluctuate daily.
Consequently, the table and graph below highlight a metric that should
only be used as a yardstick for evaluating comparative performance with
price factored in. Other architectural benefits are not covered,
obviously.
| Graphics cards | Gigabyte GeForce 8400 GS 256MiB | ATI Radeon HD 2400XT 256MiB | HIS Radeon HD 3450 Silence 256MiB | Sapphire Radeon HD 3450 512MiB |
|---|---|---|---|---|
| Actual aggregate marks at 1,024x768 | 98.71 | 119.35 | 92.33 | 98.41 |
| Aggregate marks, normalised*, at 1,024x768 | 65.7 | 99.48 | 62.89 | 71.41 |
| Current pricing, including VAT | £25.84 | £33.26 | £31.02 | £29.95 |
| HEXUS.bang4buck score at 1,024x768 | 2.54 | 2.99 | 2.03 | 2.38 |
| Acceptable frame rate (av. 60fps) at 1,024x768 | No (all games) | No (all games) | No (all games) | No (all games) |
* The normalisation refers to taking playable frame rate into account. Should a card benchmark at over 60 frames per second in any one game, the extra fps count as half. Similarly, should a card benchmark lower, say at 40fps, we deduct half the difference from its average frame rate and the desired 60fps, giving it a bang4buck score of 30 marks. The minimum allowable frame rate is 20fps but that scores zero.
As an example, should a card score 120fps we treat it as 90fps as only half the frame rate above 60fps is counted for the bang4buck - this is the formula: (120-((120-60)/2)). Similarly, should it score 30fps, we count it as only 15fps: (30+((30-60)/2)).
The reasoning behind such calculation lies with playable frame rates.
Should card A score 110fps in a benchmark and card B 160, then card B would otherwise receive an extra 50 marks in our bang4buck assessment, even though both cards produce perfectly playable frame rates and anything above 60fps is a bonus and not a necessity for most.
Similarly, without our adjustments, the aggregated bang4buck total for two very different cards would be identical if, in a further benchmark, card A scored a smooth 70fps and card B an unplayable 20fps. Both would win marks totally 180, yet the games-playing experience would be vastly different.
A more realistic (and useful) assessment would say that card A is better because it ran smoothly in both games - and that view would be accurately reflected in our adjusted aggregation, where card A would receive 150 marks (85+65) and card B 100 (100+0).
In effect, we're including a desired average frame rate, in this case 60, and penalising lower performance while giving frame rates higher than 60fps only half as much credit as those up to 60fps. If this doesn't make sense or you have issue with it, please hit the HEXUS community.
Here's the HEXUS.bang4buck graph at 1,024x768.
.
Let's be abundantly clear, the HEXUS.bang4buck graph simply takes games' performance and pricing into account. The pricing excludes shipping and handling charges, too.
The comparison Radeon HD 2400 XT GDDR3 is the most-expensive card, at £33, but its normalised performance is substantially higher than the rest, leading to the highest HEXUS.bang4buck score.
There are no stinkers here, however.
Temperature musings
| Graphics cards | Gigabyte GeForce 8400 GS 256MiB | ATI Radeon HD 2400XT 256MiB | HIS Radeon HD 3450 Silence 256MiB | Sapphire Radeon HD 3450 512MiB |
|---|---|---|---|---|
| Ambient temperature | 18.5°C | 21°C | 18°C | 17.5°C |
| Idle temperature | 47°C | 32°C | 39°C | 32°C |
| Load temperature | 60°C | 45°C | 51°C | 39°C |
| Ambient-to-load delta | 41.5°C | 24°C | 33°C | 21.5°C |
A 120mm fan was blowing over each card in our test-rigs.
The GeForce 8400 GS runs hottest out of the three passive cards. The Sapphire's heatsink works particularly well, yielding under-load temperature of just 39°C. As such, it's difficult to determine why HIS' larger heatsink doesn't fare as well.
High-definition decode
We ran three of the cards through the first two minutes of the frenetically-paced chapter 23 of Pirates Of The Caribbean: At World's End - which is encoded in H.264 and has an average bitrate of around 32MBit/s - at both 720p and 1080p resolutions.| Graphics cards | Gigabyte GeForce 8400 GS 256MiB | ATI Radeon HD 2400XT 256MiB | Sapphire Radeon HD 3450 512MiB |
|---|---|---|---|
| Average CPU utilisation % at 720p | 17.95 | 16.32 | 17.06 |
| Average CPU utilisation % at 1080p | N/A | 17.46 | 17.01 |
The Gigabyte card cannot render out to the Dell dual-link panel and thus the 1080p results aren't included. We'll add these as soon as it's retested on a WUXGA (single-link DVI) screen.
The cards internally decode at 1080p, and that's why there's scan difference between output results for both Radeon-based cards.
These ~£30 cards are accomplished enough at hardware-assisted decode to leave the CPU relatively free, then.
Overclocking.
When framerates really begin to matter, as they do with low-end
discrete cards, overclocking to safe tolerances makes sense.We managed to raise the Gigabyte GeForce 8400 GS clocks to 555/1,110/980MHz, the HIS' to 709/709/882MHz, and the Sapphire's to 709/709/1,026MHz.
Enemy Territory: Quake Wars was then re-run at 1,280x1,024, as shown on page 11, with the overclocked cards, and the percentage performance increase noted.
| Graphics cards | Gigabyte GeForce 8400 GS 256MiB | HIS Radeon HD 3450 Silence 256MiB | Sapphire Radeon HD 3450 512MiB |
|---|---|---|---|
| Clocks | 459/918/800 | 594/594/792 | 594/594/990 |
| Core/shader increase % | 20.9 | 19.3 | 19.3 |
| Memory increase % | 22.5 | 11.4 | 3.6 |
| ET: QW 1,280x1,024 increase % | 23.8 | 14.1 | 13.3 |
The Gigabyte GeForce 8400 GS sample had excellent headroom on the core (with linked shaders) and the memory. Both Radeon HD 3450 cards overclocked equally well on the core/shaders, but the Sapphire had very little wiggle room for its 512MiB memory, which is a touch surprising given that the heatsink cools the memory chips.
The upshot is the best real-world benefit is derived from the 8400 GS. Remember, of course, that each sample may overclock differently.
