AMD’s Radeon HD 4830 graphics processor

Ok, folks, this is gonna be a quick one. AMD has a brand-new Radeon to unveil today, and it’s certainly worthy of our attention. However, Damage Labs is humming away with the sound of a great many things being tested right about now, so our time to devote to this new graphics card is limited. We’ll be in and out of our look at the Radeon HD 4830 in no time, faster and cleaner than a celebrity marriage.

Little bro steps out
Yep, this new card is indeed called the Radeon HD 4830. The name tells you almost everything you need to know about this product, which would appear to be the last piece of AMD’s 4000-series Radeon lineup to fall into place. Those of you who read our recent review of affordable graphics cards may recall that AMD didn’t have much to offer between the (sensational for its price) Radeon HD 4670 at 80 bucks and the all-world Radeon HD 4850 at about $180. Well, that’s where the 4830 comes in.

This new model is, like the 4850 and 4870, based on RV770 silicon, but in its tamest form yet. Yes, folks, the great product segmentation game continues with yet another chip having perfectly good—or possibly totally flawed—bits and pieces deactivated to maintain a neat separation between models. On the 4830, two of the RV770’s 10 SIMD units have been disabled, reducing shader power (and likely performance) somewhat. Since those SIMD units are tied to texture management units, the GPU’s TMU count has dropped proportionately. The end result: the Radeon HD 4830 has a total of 128 shader execution units—or 640 stream processors, in AMD parlance—and can filter up to 32 textures per clock.

That’s it for the neutering, though. The 4830 keeps all four of the RV770’s render back-ends and associated memory controllers intact, leaving it with an aggregate 256-bit memory interface. The card’s GPU core runs at 575MHz, and it comes with 512MB of GDDR3 memory clocked at 900MHz (or 1800MT/s, for those of you keeping score at home.)

All of those numbers, acronyms, and GPU gobbledygook should add up to a pretty competitive product at its price. But, to give your brain a rest, here’s a nice picture.

Our sample card, which came from AMD, is based on the same basic board design as the Radeon HD 4850. The two are practically visually indistinguishable, right down to the single-slot cooler. However, AMD says board vendors will likely ship Radeon HD 4830 cards that use custom board designs and custom coolers, some of which will likely be of the dual-slot variety.

Somewhat unexpectedly, the 4830 also shares the 4850’s board power rating of 110W, even with its lower clock speeds and shader-ectomy. Why? AMD says it’s because 4830 cards may include chips that didn’t quite pass muster for use in the Radeon HD 4850 or 4870. Those GPUs may need a little extra juice (that is, voltage) in order to do their thing. However, as you’ll see, at least our copy of the 4830 didn’t draw nearly as much power as our 4850.

The final bit of information you need to know about the 4830 before we move on to our performance testing? Pricing, of course. AMD’s suggested “e-tail” price is $129, smack-dab in the middle of the hole in the Radeon HD 4000 lineup. That puts the 4830 almost directly opposite the GeForce 9800 GT, and AMD identifies that card as the 4830’s most direct competitor.

Such things are never entirely straightforward, though, these days. Nvidia points not to the regular ol’ 9800 GT but the higher-clocked variants like this MSI card for $119.99 at Newegg as the 4830’s truest competition. The MSI’s core clock runs at 680MHz, well above the 9800 GT’s 600MHz baseline speed. Right now, that card packs a $20 rebate, as well, potentially improving its value proposition, if you’re willing to risk seeing your 20 bucks ground up in the teeth of a bureaucracy designed to minimize redemption rates because you didn’t write neatly enough on the little form.

Ah, I love rebates.

AMD can play this game, too, of course. Right here, you’ll find a 4830 card from Sapphire for $129.99 with a $10 mail-in rebate attached. So, at the time of publication, the 4830’s net price is a little higher than a hot-clocked 9800 GT, but that could change overnight. Prices could drop or rise, and rebates could expand or evaporate. Radeon board makers could intro 4830 variants with higher clock speeds, as well, or a killer new S3 Graphics product could turn the market on its ear. So who knows? Let’s just look at how the Radeon HD 4830 performs, and I’ll leave the fine-tuned deal-mongering to you.

Before we go on, though, I should mention up front that the GeForce 9800 GT card we’ve tested in the following pages is not a higher-clocked card. Instead, it’s a Palit card with a bone-stock clock speed and 1GB of memory. That extra memory isn’t likely to do much of anything for the 9800 GT at the resolutions we’ve tested, but higher clock speeds surely would—something to keep in mind.

Our testing methods
As ever, we did our best to deliver clean benchmark numbers. Tests were run at least three times, and the results were averaged.

Our test systems were configured like so:

Thanks to Corsair for providing us with memory for our testing. Their quality, service, and support are easily superior to no-name DIMMs.

Our test systems were powered by PC Power & Cooling Silencer 750W power supply units. The Silencer 750W was a runaway Editor’s Choice winner in our epic 11-way power supply roundup, so it seemed like a fitting choice for our test rigs.

Unless otherwise specified, image quality settings for the graphics cards were left at the control panel defaults. Vertical refresh sync (vsync) was disabled for all tests.

We used the following versions of our test applications:

• Call of Duty 4: Modern Warfare 1.7
• Crysis Warhead
• Half-Life 2 Episode Two
• Enemy Territory: Quake Wars 1.5
• 3DMark Vantage 1.0.1
• FRAPS 2.9.4

The tests and methods we employ are generally publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

Specs and synthetics
We’ll get our customary start with a look at specs and some synthetic benchmarks designed to test them. The thing you need to know about the numbers below is that they come from the actual cards we tested. In some cases, those cards ran at clock speeds somewhat higher or lower than the official reference clocks established by the GPU makers, which may explain why our numbers sometimes vary from the official specifications.

	Peak pixel fill rate (Gpixels/s)	Peak bilinear texel filtering rate (Gtexels/s)	Peak bilinear FP16 texel filtering rate (Gtexels/s)	Peak memory bandwidth (GB/s)	Peak shader arithmetic (GFLOPS)
					Single-issue	Dual-issue
GeForce 9500 GT	4.4	8.8	4.4	25.6	90	134
GeForce 9600 GSO	6.7	26.6	13.3	38.5	259	389
GeForce 9600 GT	11.6	23.2	11.6	62.2	237	355
GeForce 9800 GT	9.6	33.6	16.8	57.6	339	508
GeForce 9800 GTX+	11.8	47.2	23.6	70.4	470	705
GeForce 9800 GX2	19.2	76.8	38.4	128.0	768	1152
GeForce GTX 260	16.1	36.9	18.4	111.9	477	715
GeForce GTX 260 216 SPs	18.1	46.7	23.3	117.9	607	910
GeForce GTX 280	19.3	48.2	24.1	141.7	622	933
Radeon HD 4650	4.8	19.2	9.6	16.0	384	–
Radeon HD 4670	6.0	24.0	12.0	32.0	480	–
Radeon HD 3850	11.6	11.6	11.6	57.6	464	–
Radeon HD 4830	9.2	18.4	9.2	57.6	736	–
Radeon HD 4850	10.0	25.0	12.5	63.6	1000	–
Radeon HD 4870	12.0	30.0	15.0	115.2	1200	–
Radeon HD 4870 X2	24.0	60.0	30.0	230.4	2400	–

On paper, the 4830 matches up well against the GeForce 9800 GT. The two share the exact same memory bus width and clock speed, so they have the same peak theoretical memory bandwidth. The 9800 GT would appear to have quite a bit more texturing power, and the 4830 seems to have a pronounced edge in shader arithmetic rates.

In practice, the picture is a little different. The Radeon HD 4830 takes both the color and fill rate tests and three out of the four shader tests, although its margins of victory in the shader tests aren’t as resounding as the gigaflops numbers might suggest.

Call of Duty 4: Modern Warfare
We tested Call of Duty 4 by recording a custom demo of a multiplayer gaming session and playing it back using the game’s timedemo capability. We’ve chosen to test at display resolutions of 1280×1024, 1680×1050, and 1920×1200, which were the three most popular resolutions in our hardware survey. We generally enabled image quality enhancements like 4X antialiasing and 16X anisotropic filtering, and we also added 2560×1600 to the list for the very fastest cards, in order to really stress them. For the slower cards, we also tested at 1280×1024 with antialiasing disabled, as well.

This is a close one, but the Radeon HD 4830 is slightly faster than the 9800 GT, especially at 1280×1024 where the GPU is less of a performance constraint, possibly because AMD’s graphics driver executes a little more quickly.

The big thing to take away from this test is simple: either card will run this game very acceptably at up to 1680×1050 resolution. At 1920×1200, you may have to cut back on image quality options like antialiasing a bit, especially in multiplayer.

Half-Life 2: Episode Two
We used a custom-recorded timedemo for this game, as well. We tested with most of Episode Two‘s in-game image quality options turned up, including HDR lighting. Reflections were set to “reflect world,” and motion blur was disabled.

This is a clear win for the Radeon HD 4830, obviously. Then again, even the Radeon HD 4670 can run this game reasonably well at 1920×1200 with these image quality settings (which are quite good).

Enemy Territory: Quake Wars
We tested this game with “high” settings for all of the game’s quality options except “Shader level” which was set to “Ultra.” Shadows and smooth foliage were enabled, but soft particles were disabled. Again, we used a custom timedemo recorded for use in this review.

Chalk up another one for the 4830. We’re still talking about frame rate averages of over 60 FPS at 1920×1200, but the 4830 practically shadows the GeForce 9800 GTX+ here, leaving the 9800 GT in its dust.

Crysis Warhead
Rather than use a timedemo, I tested Crysis Warhead by playing the game and using FRAPS to record frame rates. Because this way of doing things can introduce a lot of variation from one run to the next, I tested each card in five 60-second gameplay sessions. The benefit of testing in this way is that we get more info about exactly how the cards performed, including low frame rate numbers and frame-by-frame performance data. The frame-by-frame info for each card was taken from a single, hopefully representative play-testing session.

We used Warhead‘s “Mainstream” quality level for testing, which is the second option on a ladder that has four steps. The “Gamer” and “Enthusiast” settings are both higher quality levels.

Obviously, this game puts quite a bit more strain on the GPU. The 4830 handles it with composure, basically matching the GeForce 9800 GT. With frame rate minimums around 25-28 FPS, both cards run Warhead quite nicely at these settings. I enjoyed blowing stuff up with each.

Power consumption
We measured total system power consumption at the wall socket using an Extech power analyzer model 380803. The monitor was plugged into a separate outlet, so its power draw was not part of our measurement. The cards were plugged into a motherboard on an open test bench.

The idle measurements were taken at the Windows Vista desktop with the Aero theme enabled. The cards were tested under load running Half-Life 2 Episode Two at 1680×1050 resolution, using the same settings we did for performance testing.

The 4830’s power consumption is admirably low given this card’s performance. The 9800 GT draws more power at idle and when running Episode Two. This is one area where a higher-clocked version of the GeForce 9800 GT would not help even the score, either.

Noise levels
We measured noise levels on our test systems, sitting on an open test bench, using an Extech model 407727 digital sound level meter. The meter was mounted on a tripod approximately 12″ from the test system at a height even with the top of the video card. We used the OSHA-standard weighting and speed for these measurements.

You can think of these noise level measurements much like our system power consumption tests, because the entire systems’ noise levels were measured. Of course, noise levels will vary greatly in the real world along with the acoustic properties of the PC enclosure used, whether the enclosure provides adequate cooling to avoid a card’s highest fan speeds, placement of the enclosure in the room, and a whole range of other variables. These results should give a reasonably good picture of comparative fan noise, though.

Well, OK, the 4830 and many of the other cards were quiet enough not to register on our sound level meter, which doesn’t go below ~40 dB. What this result tells you is that the 4830—and many of these other cards—were nice and quiet for us. In other news, I have a nice, new sound level meter sitting here that’s capable of measuring noise levels down to ~26 dB. We’ll have to try it out soon.

GPU temperatures
Per your requests, I’ve added GPU temperature readings to our results. I captured these using AMD’s Catalyst Control Center and Nvidia’s nTune Monitor, so we’re basically relying on the cards to report their temperatures properly. These temperatures were recorded while running the “rthdribl” demo in a window.

85°C seems to be par for the course for the Radeon HD 4800 series. AMD insists these temperatures are well within its expectations, and I can live with that. Just don’t try to pull a hot card out of your system without letting it sit for a while, or your fingers are in for some pain.