GeForce G310 Fermi - Nvidia news! 2010

An overview about the performance

The initial performance of the first card based on Fermi gave a small possibility of seeing the performances that the card can offer. The frequencies and specifications aren’t known yet, and the very small session consisted of the integrated benchmark on Far Cry 2.

The choice of using a title such as Far Cry 2 wasn’t casual: it’s one of the few titles where the ATI Radeon HD 5000 cards do not reach excellent performances with the anti-aliasing filter on. The values we’ll be giving next were supplied by nVIDIA, and the hardware system used by nVIDIA for the test included an Intel Core i7 960 CPU, 6 DDR3, triple-channel memories, hence results that can be easily achieved without too much hassle.

Far Cry 2 - Ultra Quality / AA 4x / DX10

Resolution: 1920 x 1200

- Average FPS: 84.02
- Minimum FPS: 65.1

The values are to be taken lightly, as there are no information about which drivers or game patches were used. But it’s good for reference, until we can get our hands in a real card.

In fact, the usage of two cards in SLI is not only a necessity in the case of the 3D Surround Gaming, but also highly recommended: the performances are very reduced when using three screens in parallel, while the 3D technology puts yet another strain on the frames per second generated by the card. Having two GPUS operating in SLI makes the impact of the 3 screens and the 3D technology a lot smaller, and allowing users to have a decent playability.

A technical note before passing to the next point: the surround technology allows users to have 3 different monitors connected to the system, but the 3D Vision Surround needs three identical monitors: managing a 3D signal means that the monitors should be able to manage a 120Hz refresh, and using the same monitors would help avoid timing problems.

ATI hasn’t yet implemented the CrossFireX support to the Eyefinity technology; consequently, it’s not yet possible to take advantage of more GPUs in the moment that games are executed in more than one screen with surround gaming, and thus not using all the power the system has in that case. Perhaps with future drivers, ATI will overcome this limitations. nVIDIA has done so already, indeed, it’s not possible to have the surround gaming implemented without the SLI technology at all.

This makes us think that perhaps the 3D Vision Surround technology was developed by nVIDIA as a proper response to the Eyefinity from ATI. Two points lead to that conclusion: one, the necessity of two cards in SLI with the GTX 200 cards, and the lack of support for more than two displays on the GF100 for each GPU.

In order to make the surround technology work, nVIDIA requires at least two cards connected through SLI technology. The compatibility reaches the GeForce GTX 200 cards, based thus on the GT200 GPU, besides those with Fermi architecture, of course.

The need for two cards is a practical one: each GTX 200 card (and the same goes for the Fermi cards) can manage only up to 2 simultaneous screens, despite the fact that the number of display connectors could be higher than that. The presence of two cards connected through SLI allows the cards to manage the 3 displays, and also benefits from two GPUs for all the 3D renderings.

It’s easy to make a comparison with the Eyefinity technology from ATI. Eyefinity allows the card to manage up to six different monitors at the same time. The current approach on the market, featured on the Radeon HD 5000 cards allows for three monitors put together horizontally, making it so the game recognizes the three screens as just one panel.

The strategy used by nVIDIA with the 3D Vision Surround is indeed similar to the Eyefinity technology, except it is limited to 3 screens, while in theory, Eyefinity can manage up to 6 (not on the current models, however). It’s also posssible to enable the surround mode without the 3D Vision mode, for those who don’t want to enable the 3D feature with a specific title, for example.

3D Vision Surround

One of the new features that will be implemented with the GeForce cards based on GF100 architecture is called 3D Vision Surround, shown at the CES in Las Vegas. As the name says, the 3D Vision Surround technology allows the card to use the 3D Vision technology on three monitors, one next to the other, in a “surround” mode. The result is an extended gaming surface that gives more immersion than what would happen on a single screen.

Putting three monitors together greatly extends the horizontal resolution, with a direct impact on the overall performances: the card should, indeed, generate three times more pixels compared to a single monitor. Such a technology requires a big elaboration power, and a correct balance in the type of screens to be used: in this case, the ideal resolution would be 1920×1080 pixels (or 1920×1200 pixels, depending on what monitor is used), obtaining a resolution of 5760 pixels with three monitors.

The fact that the GeForce GF100 has high energy consumptions isn’t that much of a mystery: it’s easy to determine that by the fact that each card is connected with two PCI-Express connectors, a 6-pin one and a 8-pin one.

The Radeon HD 5800 series cards feature two 6-pin connectors, while only the dual-GPU HD 5970 has the same connectors as the GF100. It’s possible that these connectors will be different on the final version of the card, but we think it’s not likely to be that way.

The cooling system should be similar to what was seen on the single-GPU GTX 200 family: hence, a fan that brings the air from the inside and cools the radiator directly in contact with both the GPU and the GDDR5 memories, and also the power supply circuits. After the end of the cycle, the air is pushed out through the grid present above the display connectors.

nVIDIA has shown many demonstrations during the CES in Las Vegas, on systems based on GeForce Fermi cards, with single-GPU, SLI and Triple SLI configurations. All the Triple SLI configurations were based on a liquid cooling system as it can be shown on the picture above. We cannot know for sure if the liquid system was chosen due to aesthetic reasons or due to the heat generated by the cards in triple-SLI.

Starting from the card’s size: the length seems to be similar to those that can be found on the GeForce GTX 200 currently present on the market, and that shouldn’t pose any problems in the installation on most part of the cases in the market. In this sense, nVIDIA has decided to keep a certain standard instead of following ATI’s steps that brought better performing cards with their HD 5800 series, but longer and occupying more space.

The design also sees a heatsink that occupies two slots, covering the entire PCB and leaving the backside uncovered. The uncovered area has no meaningful components such as memory modules. The 384-bit bandwidth requires 12 GDDR5 memory models, each one of them connected to the memory controller with 32bit bus.

The Card

Similar to what happened back in October, when during their GPU Technology Conference, nVIDIA made the first public demonstration of a working Fermi solution within a demo system but showing a non-working dummy board to the participants, nVIDIA hasn’t shown the current GeForce prototype that will be based on the GF100, even if this model could be considered to be very similar to the card that will probably be launched from March on.

There are only a few images of the real card used, thanks to which we can make some considerations on the architecture and the choices made by nVIDIA. These pictures were taken during the CES 2010, during stand presentations and press meetings.