Visceral Games has said that Dante's Inferno is a better-looking console game because it runs at 60 frames per second, as opposed to more standard 30FPS.

In a wide-ranging interview with Gamasutra, creative director Jonathan Knight explained how the team had "a fair amount of angst" over the decision, but worked hard in evangelising the call to lock at 60 and getting the development team on-board.

"I think any artist would be lying if they said that they didn't prefer to have more bandwidth," he said. "Any milliseconds you give them, they're going to use it on just one more effect, or what-have-you. But what we found is, it's more of a question of willpower than a technology question. And you just have to commit to it, and say, 'Here are your budgets. Here's the box we're gonna play in.'

"30 frames is a very challenging box to play in as well, and so once you just get everybody bought into that, then what I've found is that the visual effects artists, and the environment artists, and so forth, they just found ways to make stuff look good at 60, and you just have to hold them to it."

Knight also believes that smoother motion helps improve the quality of the graphics:

"If you were to take a screenshot, you might be able to point out, like, 'OK, here's the compromise you made because of your frame-rate,' but when you sit and play the game, the overall visual experience is enhanced by the fast frame-rate. So, I can't really decouple graphics from frame-rate; I don't feel like it's an either/or situation."

The Dante's Inferno performance level certainly is impressive. In Digital Foundry tests on the demo code, the game performed at an almost completely uniform 60FPS in all gameplay situations, while in the recent Face-Off Dante's Inferno proved to be effectively identical in both visuals and frame-rate on PS3 and Xbox 360

Knight's comments are somewhat at odds with those made by Mike Acton of Insomniac, who has committed the PS3 developer to "the best-looking games", inferring that the more standard 30FPS is the best way forward. Our take is that it all depends on the game, a view shared by Turn Ten's Dan Greenawalt in an interview this week with VG247.

"It really turns into a trade-off for the type of visual direction you're trying to achieve, and with Forza Motorsport, the visual style is clear, crisp graphics, and highly-detailed environments and textures," he said. "But in truth, the decision is less about graphics than feel. We prioritise 60FPS as an important feature because it gives the games a feel you just can't achieve at 30; for instance, the responsiveness and feedback of the controls, or physics calculations and visual manifestation of that on how we model our tyre flex and body roll.

"While those calculations are decoupled from the graphics and run as high as 360 frames per second, we found that graphical frame-rate impacts the feel of those systems as well. Simply put, we couldn't have achieved the experience we wanted if the game only ran at 30FPS. I'm sure you would get a similar response from Infinity Ward regarding Call of Duty."

Greenawalt also agreed that smoother motion can produce better-looking visuals.

"The reality is that racing games running at 30FPS have to deal with visual shuttering artifacts in the environment and backgrounds that fly by your field-of-view when the car is traversing at high speeds. So you then go and mask that using motion blur for your environments, which end up eating into your GPU cycles, which take up resources from other features you want on-track."

THQ has released a patch said to address the screen-tear issues in the Xbox 360 version of Darksiders, with the code auto-updating the next time you run the game while connected to Xbox Live.

Earlier in January, developer Vigil Games said said that "it's something that's really not a difficult fix for us" while sources at THQ mentioned that the problem didn't appear in preview versions of the game - suggesting some kind of issue that materialised only in the run-up to launch.

It's extremely rare that patches actually increase the performance level of a game. Cutting down screen-tear to an appreciable degree without impacting frame-rate involves some pretty substantial optimisations to the engine, so it was with some level of intrigue that Darksiders performance was put through the Digital Foundry frame analysis tools in both original and patched conditions.

The result? Well, screen-tear is indeed dramatically reduced. In the sample of clips below, the launch version of the game has 43.5 per cent torn frames. The new version is considerably improved, taking that number down to 21.6 per cent.

Tearing from the old version of the game is on the top of the graph, while performance of the new patch is below. The green graph line indicates frame-rate from the launch version, while cyan shows the update.View this video on EGTV

So, what happened? How has the tearing been reduced by a factor of over 50 per cent? Based on what the video above shows us, it looks to be more of a case of the original code running with no limit to frame-rate. The game seeks to pump out as many frames as possible (maxing out close to 60FPS in some places), resulting in more tearing. The patch seems to be capping maximum output from the engine at 30FPS, thus giving the game engine a bit more room to breathe in less strenuous scenes.

While the video isn't exactly like-for-like (though the comparison clips are taken from the same game areas, synced to the same cut-scenes), it does suggest that when performance dips below 30FPS, both the launch and patched versions seem to act in much the same way. In short, it's fair to say that capping frame-rate in this way produces a more stable look to the game. So, how does this stack up to the PS3 version of Darksiders? It's one of the games currently being analysed for the next Face-Off, so look out for that next week...

It's the must-have download this Christmas. Polyphony Digital is on the brink of launching a brand new Gran Turismo 5 demo, and we've got an early preview. Already its treasures have been plundered with a 1080p screenshot gallery and an HD edit showing off the intro screen, gameplay and replay quality. Now it's time to put the code up to the usual DF image quality and performance tests.

First things first, there has been some talk about this sampler not actually being a GT5 demo. This is true in the sense that it is a one-circuit, lap-time-driven time trial: an extremely small, very limited and ultra cut-down version of what promises to be one of the most content-rich games seen on console. However, the wording on the splash screen itself confirms that - in the eyes of Polyphony Digital at least - this is indeed a "GT5 demo", representative of the driving experience of the full game.

The demo's splash screen confirms that this is indeed a GT5 demo, though content and options are limited for the purposes of the time trial competition.

In terms of actual rendering spec, not much has changed at all from the GT5 Prologue code. Limit your PS3 to 720p mode and you get native resolution with 4x multi-sampling anti-aliasing (MSAA). Invoke the might of 1080p, and the framebuffer is set at 1280x1080. That's a 50 per cent resolution increase over 720p mode, with AA dropped to 2x. The RSX scaler then presumably kicks in with its horizontal bilinear resize to 1920x1080, and the HUD elements are overlaid on top of that. While GT5P ran its menu screens at full 1080p, it appears that the new demo's intro screens stick with the same resolution as gameplay.

Rendering tech looks mostly unchanged from the GT5 Prologue. Top shots give examples of the 720p mode in action, while the bottom shots showcase the 1080p display.

Content-wise, there's just the one track - the Indianapolis racing circuit - and a mere brace of cars. Well, just the one really: the Nissan 370Z, but selectable in both road-going and racing-spec. The various race options found in GT5 Prologue, including driver-aid selection, are off-limits in order to make sure that everyone faces the same driving challenge. This is an online time trial, after all, and times need to be like-for-like.

For those not used to the simulation settings of Gran Turismo, this makes the two vehicles included in the demo something of a handful. The road version of the 370Z in particular feels like an absolute lump, while keeping the rear end of the race-spec version under control takes some doing, especially if you've conditioned yourself to Forza Motorsport 3's arcade handling options in recent months.

So, in terms of stuff to render at any given point, this GT5 demo is fairly simplistic: just a single car (two if you count the ghost), and a fairly unambitious track. In theory this should be a walk in the park for the GT5 engine - the very best conditions for the engine to shine. Two performance tests were in order. We ran laps in both in-car and external views, and we also ran them both at 720p and 1080p. Where possible we tried to keep the ghost vehicle in our sights as much as possible to add that extra level of stress to the engine. First up, the 720p performance test:

Performance analysis of the game running in 720p mode.View this video in HD

Around five per cent of the total 60Hz output consisted of torn frames, with a minimum frame-rate of 52FPS. It's interesting to note that the combination of rendering the fence and the ghost car simultaneously appears to have the most impact on frame-rate, but it is a touch concerning that such a basic-looking track with just one other car in view should affect performance at all. Switching to 1080p, the same test was re-run.

GT5's 1080p mode really is very impressive, with only a small performance penalty over the 720p mode, which isn't noticeable at all within this demo.View this video in HD

Results were fairly similar: the same low of 52FPS, though the torn frame count elevated to around 12 per cent of the system's output. Overall, this seems to be in keeping with the approximate performance level that we saw in GT5 Prologue in much the same conditions. Whether it will be improved in the final game out next year remains to be seen, though it should be pointed out that the gamescom demo we saw earlier in the year also ran in a very similar fashion to Prologue.

It's interesting to note Polyphony's approach to the ghost lap. Rather than render the car as a transparency, an interlaced approach is used in order to lower bandwidth and fill-rate. Maintaining 60FPS is a tough call and while Forza 3 is probably the better example of a game that cuts down its alpha textures to an absolute minimum, GT5 too uses similar tricks in order to maintain frame-rate. Alpha-to-coverage (a cheaper transparency technique) is used in both games, though the use of MSAA does mitigate the screen-door effect.

As you would expect from a Polyphony Digital game, overall image quality is outstanding, although it is rather odd that the low-quality shadow filtering should look quite as jaggy as it does considering that (certainly in 720p mode), GT has plenty of MSAA to make things look smoother.

Content-wise, obviously the demo is rather limited. It is what it is: a one-circuit time trial. In that respect, it works rather well. Racing against your own best lap-time via the provided ghost driver is great fun and having Kazunori Yamauchi's personal best on-screen at every given moment offers plenty of encouragement to refine your skills.

Visceral Games' Dante's Inferno has received its first public outing with the release of a playable demo that's available to download from the American PlayStation Store. First impressions are that the game is clearly inspired by Sony Santa Monica's epic God of War series, albeit with the ancient Greek mythology swapped out for Crusader-era Christianity, as the eponymous Dante battles enemies of a literally biblical scale, Kratos-style.

The DF video analysis of this particular demo is actually much more useful as a preview of the demo itself, as the FPS analysis is what you might call rather one-note. Frame-rate graphs are rather uninteresting as this is a 60FPS game in the truest sense of the word. Six of the eight clips selected for the video are from gameplay, and out of the 25,000+ put through our tests we only saw a measly two dropped frames: 0.008 per cent of the overall capture - of course completely invisible to the human eye. Tests confirm the drop to a locked 30FPS for the cinematics.

A fair few games aspire to sustained, uninterrupted 60FPS gameplay, but few achieve it. Dante's Inferno looks like being one of that select group.View this video in HD

The truly impressive performance we see here does come with a cost of course. Compared to its immediate competitor God of War III the geometry, detail and effects are significantly pared back. Resolution is confirmed as native 720p, but there is no anti-aliasing. On the plus side, this is 60FPS with no apparent compromise: the game is v-synced and absolutely rock-solid, certainly in the sampler levels provided in the 1GB download.

Where the game is a touch on the disappointing side is in its cinematics. The drop to 30FPS isn't a particular concern, but while it's clear to see that elements such as the player model gain detail, there's still no anti-aliasing and no motion blur or other post-processing effects, resulting in a sometimes-rough appearance.

More than that, the game switches between FMV sequences and engine-driven cut-scenes, making for a rather jarring jump in image quality. The inclusion of high-end CG and even what you might call cartoon animation mean the overall presentation is somewhat inconsistent.

Overall though, first impressions are favourable. In an industry where the 30FPS is effectively the standard, it's refreshing to see Visceral Games sustain 60FPS so impressively, especially in a cross-platform project. Yes, Dante's Inferno will also appear on Xbox 360, but the corresponding demo won't be appearing on Xbox Live until Christmas Eve, when we'll all be on holiday. If we can sneak a look at the 360 demo during the festivities, we'll see about updating the DF Twitter feed with our thoughts.

Aside from some blurry pre-alpha screenshots and unconfirmed game details, next to nothing has been revealed about Bungie's 2010 epic - Halo: Reach. Until now. Over the weekend, a new video was released, showcasing a massive technological leap over the previous Halo engine.

So, just how "real" is the new trailer? Prior to its unveiling, wording on the content was carefully chosen. The video would be "in-engine", the key phase missing there being "real-time". It's fairly obvious that what we have here is a offline render of a Reach cut-scene: Bungie effectively allowing the engine all the time it needs to create a massively high-resolution version of the cinematic, which is then downscaled to native 720p, eliminating all aliasing and producing a superb-looking presentation.

What we're seeing here is indeed in-game assets running in the new engine, it's just that we have no real idea how this will translate to actual run-time performance. More aliasing is an obvious given, but it's unclear what other tweakables within the engine Bungie ramped up in order to produce the best-looking trailer.

A high-quality, super-streamable, hand-encoded version of the Halo: Reach trailer, presented for your viewing pleasure.View this video in HD

Back at E3 2006, two versions of the Halo 3 trailer made it out in the wild: a full-on super-sampled, 60FPS video, along with a run-time 30FPS rendition that exhibited plenty of aliasing. In that case, Bungie was sampling the code on the showfloor, so we'd be hugely surprised if a corresponding run-time version of this video were released.

What we do see in the trailer offers up plenty of cool discussion points though. Models, geometry and textures certainly look as though they are based on in-game assets, and in-game atmospherics and lighting are phenomenal. Material shaders in terms of the armour, vehicles and human faces are of an outstanding quality.

Indirect lighting (shown about a minute into the trailer) looks of a high quality and it's interesting to note that the shadows are of a high quality too. Coming out of the building, we see some low-resolution shadows, but the effect is decent owing to high-quality filtering. Human characters in terms of artwork and especially animation are hugely improved over the somewhat-average efforts in Halo 3 and ODST. Bungie also has a state-of-the-art motion capture studio in-house now that should help improve animation still further.

Interesting artefacts in the trailer - first up (left), a depth-of-field bug where only the bottom half of the wing in the foreground is processed. In the second shot (right), note the low resolution shadow cast by the door. It still looks fine though owing to a large level of filtering. Aspects like this show this is indeed in-engine as opposed to pre-rendered CG.

Over and above all of this, it is also worth remembering that the cinematics often show the engine running in a "best case" scenario. Certainly, in Halo 3, the cut-scenes often showed a higher graphical quality than the actual gameplay - the scene where the Arbiter stabs the Prophet of Truth showed a dramatic difference in lighting. How or whether this translates to Reach won't be apparent until we get to see proper in-game screenshots, which we don't expect until the beta launches some time in 2010.

With the tech stuff out of the way, does the Reach trailer give us any hints about how the new game fits into the established Halo canon? Obviously the fact that we get to see plenty of Spartan IIs in action is the big news here, but what is curious is that according to the existing mythos, there were only ever 75 of them, hand-picked from 150 children, hence John-117, the iconic Master Chief.

Spotted in the trailer were Jorge-052 (tank guy), Carter-259 (leader guy), Kat-320 (speedrunner with less armour from waist-level down). It'll be interesting to learn whether Bungie is adjusting the canon by introducing a new batch of Spartans we don't know about, or whether we are seeing the Spartan-IIIs here - cut-down, cannon-fodder troops lacking the refined genetics of the Spartan-IIs. As promised by Bungie, we also see the return of the ODSTs: the anonymous male's helmet was of the same design.

With the Halo Waypoint now established and a clear effort being made to build and reference an established timeline of events, it'll be intriguing to see how this plays out as more details on Halo: Reach emerge.

Thanks to Alex Goh for his input on this piece.

Update: This morning, Christer Ericson, director of tools and technology at Sony Santa Monica and part of the hugely respected God of War team, got in touch to set us straight on the technical background to the effect seen in The Saboteur. Christer's knowledge on this subject obviously far outstrips our own and he took issue with several points in yesterday's article, which you can still read in full below this update.

Firstly, the definition of MLAA isn't in keeping with the basic outline of the technique as it has been described by some Pandemic staff, and it is almost certainly an extension of the edge-filter plus blur technique seen in several cross-platform titles already. As Christer says, "the qualitative difference comes down to how you edge-detect and how you 'blur'."

The screenshots may not be showing MLAA, and it's almost certainly not a technique as experimental as we thought it was, but it's certainly the case that this is the most impressive form of this type of anti-aliasing we've seen to date in a console game. Certainly, as we alluded to originally, the concept of using an edge-filter/blur combination isn't new, and continues to be refined. This document by Isshiki and Kunieda published in 1999 suggested a similar technique, and, more recently, AMD's Iourcha, Yang and Pomianowski suggested a more advanced version of the same basic idea.

In terms of the effectiveness of luminance for determining edges, Christer also pointed out that the pixels we've highlighted as causing problems for edge-smoothing in The Saboteur aren't a product of red meeting black, but actually brown and red - similar in terms of luminance values and thus more likely to cause that particular artefact.

It's fair to say that in our excitement about the level of image quality displayed in The Saboteur, we were off-beam to be so definitive about MLAA as the technique being used, but it is still a great example of the SPUs being used to do post-processing work more traditionally associated with the GPU and, in this case, producing arguably better results than the more generally accepted MSAA solutions built into the graphics hardware. Thanks to Christer for getting in touch.

Original story: If there's one element of multi-platform development that comes up repeatedly in our Face-Off coverage, it's the implementation of anti-aliasing on current-generation HD consoles.

It's often the case that the Xbox 360 version of a game features edge-smoothing, while the PlayStation 3 counterpart will either leave it off altogether, scale it back, or use an NVIDIA specific technique called "quincunx", which refines edges nicely, but blurs the entire texture in the process.

None of these options are particularly attractive (though quincunx has its place in certain scenarios), and neither is blurring the entire screen, which other people do; the so-called Vaseline effect. However, recent games like Brutal Legend and Overlord II on both console platforms have tried out another technique - seeking out just the edges themselves and blurring them, leaving texture detail intact. It's better than nothing but still not actually that good.

The PS3 rendition of Pandemic's The Saboteur is different though. It's special. It's trying something new that's never been seen before on console, or indeed PC, and its results are terrific. In a best-case scenario you get edge-smoothing that is beyond the effect of 16x multi-sampling anti-aliasing, effectively delivering an effect better than the capabilities of high-end GPUs without crippling performance. Compare and contrast with Xbox 360 hardware, which tops out at 4x MSAA.

Let's kick off with a quick comparison of the effect in play on both versions of the game. It's interesting to point out that both Xbox 360 and PC versions of The Saboteur have no support for anti-aliasing whatsoever. It's just the PS3 owners that get the love, for reasons explained later. While the lack of AA in the Xbox 360 version is a bit disappointing, for the purposes this piece it's a bit of boon as it effectively gives us a "before and after" means of better understanding the PS3 technique.

The Saboteur on 360 has no anti-aliasing, making it a good comparison point up against the PS3 version with its new edge-smoothing technique. The difference in lighting, along with other matters, is something we'll cover in the next Face-Off.

So how is it done? Well, in the comments section of one US PlayStation Blog post, Pandemic's Tom French talked about "using the SPUs to do a full-screen FSAA filter". The satellite processors within the Cell chip are excellent for extremely fast processing of limited batches of data, making them perfect for the task at hand, which is to process the entire framebuffer seeking out all edges and then blending them.

Posters on the Beyond3D forum soon began to investigate. It's a technique initially put forward by Intel, but best described with examples in this blog post which shows the real potential of the technique, and how it compares with the Brutal Legend method of edge-blurring. There is no real competition here. The so-called morphological anti-aliasing (or MLAA) seen in The Saboteur is leaps and bounds ahead of anything we've seen so far when it works in optimum conditions.

Being experimental, it does have its drawbacks. When edges in-game are one pixel or less than one pixel in thickness, the edge-detection technique doesn't really work. Pandemic also subjects the entire, completed framebuffer to analysis - including the HUD elements - so there is often artefacting on the text overlays. This is probably unavoidable in this game: while the GPU starts to draw the next frame, the SPUs are busy with the AA and for that situation to take place the completed frame will need to be analysed.

Pandemic is keeping mum on the AA technique, though some tidbits have apparently leaked from the developers on the NeoGAF forum. According to these posts, the filter is applied to the luminance of any given scene. It's a really clever way to maintain speed, but on the other hand, some colours - red and black for example - have similar luminance levels, so the filter picks up the majority of edges but misses others. Additionally, in some cases there are some interesting "fuzzing" issues on edges, which without some more illumination from the coders is very difficult to explain. Perhaps it's simply the effect of processing a motion-blurred screen?

While the edge-smoothing properties are astonishing, it comes at the cost of a few artefacts as you'll see by clicking on each thumbnail. Thankfully they barely impact on the quality of the overall presentation.

Overall then, what we have in The Saboteur on PS3 is very much an experimental technique, and you get the idea that the base visuals suit the technique well. In playing the Xbox 360 version, the level of "teh jaggies" isn't exactly a major issue; the game lacks high-contrast edges, and is pretty soft in general. In this environment, the MLAA technique employed by Pandemic works beautifully and in most cases you really need to look hard for the artefacts. But they are there, and you wonder how well the technique would work on higher contrast games like Battlefield: Bad Company 2, Halo 3 or the almighty Uncharted 2 where the MLAA would really have its work cut out.

In the meantime, what we have is something that's new and genuinely exciting from a technical standpoint. We're seeing PS3 attacking a visual problem using a method that not even the most high-end GPUs are using. You can't help but wonder whether MLAA, in combination with MSAA and a filter to weed out the artefacts, couldn't be hardware-integrated in the next generation consoles.

It'll also be interesting to see whether MLAA returns in other PS3 cross-platform projects before then, because it looks extremely good in action. Chances are it will come down to how computationally expensive the technique actually is on the SPU, and how it can be refined still further, and that's where our knowledge hits a brick wall. If any ex-Pandemic staff want to share more on the MLAA implementation, feel free to get in touch...

Ubisoft's terrific Assassin's Creed II launched just yesterday and it really is something extraordinary - a game so impressive it's more than worthy of some special treatment at the hands of Digital Foundry. As discussed in this week's Face-Off, one of the key improvements made to the original Assassin's Creed engine was the addition of a full day/night cycle, and we were eager to test that implementation.

There's no better way to do this than by undertaking one of our painstaking, time-consuming, but really very cool time-lapse videos. Just a few special ingredients are required to put one of these together: the ability to go anywhere within the virtual world, a first-person view with no intrusive HUD, and finally, phenomenal amounts of patience. In truth, the biggest challenge with this one was keeping Ezio as still as possible (you might notice the odd bit of "wobble" on close objects where this wasn't possible).

What we have is a collection of 31 time-lapse locations, covering 10 days of in-game time across four cities, from the rustic charms of Florence through to the sheer beauty of Renaissance Venice. It's Assassin's Creed II as you've never seen it before and certainly won't see it elsewhere. In addition to the streaming options, there's a download version on the EGTV clickthrough, along with this mega-bandwidth version designed for playback on Xbox 360 and PS3.

10 days of in-game time from four major cities featured in Assassin's Creed II - presented via the beauty of time-lapse video. Music by ACII maestro Jesper Kyd.View this video in HD

It's intriguing to compare the virtual world Ubisoft Montreal has created with Rockstar North's incredible Liberty City, as seen in GTAIV. Both do a superb job in providing a fully immersive gameplay environment, but the two approaches - time-of-day lighting and shadowing aside - are very, very different. Rockstar has attempted a fully featured approximation of city life: a bewilderingly vast array of different AI behaviours for the characters, a full weather system, and of course implementation not just of NPCs, but also a complete range of vehicles too.

Assassin's Creed II seems to follow a different philosophy - the creation of environments that look incredible "in the moment", but don't really aspire to any lofty pretensions of creating a full-on simulation of life in 15th century Italy. Aside from their penchant to occasionally delve into your money pouch or engage you in rooftop chases, AI behaviour of the citizenry is unremarkably routine (those poor courtesans don't seem to get much "traffic"). Horses and vehicles aren't allowed within city limits. This "almost but not quite" approach also applies to the renderer too. The time-of-day lighting is quite superb, but the sky itself looks to be a simple 2D bitmap wrapped dome-like around the environment with a layer of atmospherics overlaid on top. It's an example of how compromise only really manifests when the engine is put through extraordinary scrutiny (as time-lapse video is wont to do). In-game it just works and looks great.

The core tech within Assassin's Creed II is very similar to that seen in the first game, which at the same time proves just how far ahead of the curve Ubisoft Montreal was in 2007, but makes the slightly dodgy performance stand out somewhat in the present day.

However, aside from the implementation of the day/night cycle, there have been other improvements too. The developers have rarely discussed the new "Scimitar" engine for ACII with the press, but mooted additions have included improved lighting and reflections (essential for the spectacular Venice cityscape, obviously), while the vegetation generation system pioneered in Far Cry 2 also gets ported over for inclusion within the revised tech. Ubisoft Montreal has also made headway in improving the level of detail (LOD) levels with new streaming tech that resolves higher-quality art further away in the distance.

However, the limitations of the original AC1 tech do manifest, most notably in the cascading shadowmaps within the game. Detail level of shadowing changes dramatically at set points ahead of Ezio, most noticeable in the video at around the 00:57 mark, where shadowing on the central tree is divided half and half between higher and lower resolution renderings. Noticeable geometry popping (scenery appearing from nowhere) is also another issue that is reminiscent of performance in the first Assassin's Creed and still manifests even after hard-drive installation.

Ubisoft's technical achievement here is still sizeable, but if you're thinking that the scale and diversity of this open world causes issues in terms of storage space available on Xbox 360, think again. Assassin's Creed does indeed effectively max out the available DVD space, but whereas some publishers release different game variations with support for individual languages, ACII features full audio support for English, French, German and Italian on the same disc. Each of those sound packs takes up around 200MB of space on the DVD, and over and above that, there's also another 100MB spent on a trailer for the forthcoming Ubi/James Cameron Avatar game. It's also interesting to note that the PS3 version of the game occupies pretty much the same area too.

In one sense it highlights the excellent levels of compression that Ubisoft Montreal achieved in ACII, but on the other, it does rankle slightly that one of the annoyances of the original game - repetitious dialogue from the citizenry - once again rears its ugly head in the sequel. It's an element that we would really like to see improved in the sequel.

Ubisoft is on the record as saying that Assassin's Creed will be a trilogy of games, so it will be interesting to see where the developer is heading for the next title, and whether the existing engine will stand the test of time for a third outing - presumably in 2011, when the current HD consoles will be well into their twilight years. The sheer volume of content in ACII, and the fact that the game itself is simply so utterly cool more than mitigates any of the technical shortcomings in the here-and-now. But come AC3, we can't help but wish for the complete package.

So, Call of Duty: Modern Warfare Reflex on the Wii. The basic idea of bringing one of the most advanced HD console shooters to the Nintendo platform sounds ludicrous, but Activision in concert with its Treyarch development studio has only gone ahead and done it, and the result is pleasantly surprising. Wii owners get a surprisingly fully-featured rendition, which even includes the lion's share of the online gameplay as well as the customary Wiimote-powered "waggle" controls that you won't have experienced in the initial release.

The notion of porting Modern Warfare to the Wii isn't so whacked out or as ker-azy as you might initially think for a number of reasons. The first Modern Warfare's core engine still has a fair amount of DNA left over from the Quake tech upon which the Call of Duty franchise itself was born. The basic setup of the levels in Modern Warfare offers little that any last-generation shooter couldn't cope with, and this is born out by the quality of the conversion in the Wii game.

Content-wise, it's pretty much all there. Compare and contrast with the new stages in Modern Warfare 2, streaming and decompressing on the fly from the disc - providing a vastness of scale never seen in the series before, and almost impossible to replicate on Wii without some serious re-engineering.

More than that, in some development circles, there is the school of thought that porting over your Xbox 360 game to the Wii is actually easier than it is in developing it for the PS3 (swap over the 360/PS3 roles if you like, the point remains the same). Essentially, in a market that demands platform parity in its HD console releases, the amount of effort required to achieve this is pretty exhaustive bearing in mind how different the base architecture is. In porting over to the Wii, developers face no such pressure - they have carte blanche to cut down and remove whatever they like, just so long as the final result looks okay and is playable.

Modern Warfare on the Wii. Graphically rather functional, but the last-gen elements of the core engine tech help make the transition to Wii easier to swallow.View this video on EGTV

That's exactly what Treyarch has done with Modern Warfare. Texture detail and geometry levels are savagely hacked down, lighting effects and shadowing are brutally pared back, frame-rate is cut from 60FPS down to the more manageable 30FPS (and even then it has trouble sustaining it in some of the more sustained firefights). The whole lighting model has been adjusted to such a degree that the night-vision section in The Bog is entirely superfluous - despite the game nagging you to don the goggles, the section remains so bright you don't actually need to.

Graphically, Reflex is functional, looking pretty much as you'd imagine a PS2 version to look if handled by a reasonably competent developer, but put it this way - Criterion's Black this ain't. However, stick the Wii game into a side-by-side comparison with the PS3 version and you can see that despite the vast array of cutbacks and compromises, it does a good enough job of capturing the look of the original.

A standard def head-to-head between the Wii and PS3 versions of Modern Warfare. The Wii game is serviceable enough but when Criterion's Black still looks so much better, it's clear that no technological barriers are being pushed back on the Nintendo console.View this video on EGTV

Reflex follows a very different gameplay strategy than its HD siblings. Infinity Ward prides itself on ultra-low latency controller feedback and targets 60FPS gameplay to achieve that. In the drop down to 30FPS, that feedback has been compromised, but then again, the IW control system itself is history.

In swapping out the joypad for the remote (no MotionPlus support, sorry), the lack of 60FPS doesn't really matter bearing in mind the additional "latency" of dropping the controller-based system for the use of the remote instead. Moving your arm about incurs much more "lag" than using your fingers and thumbs on the tried-and-tested joypad - the trade-off being that pointing with the Wiimote should feel more instinctual, and you should be able to target the enemy more quickly.

The Wii's waggle system operates on a similar basis to the scheme found in The Conduit and there are moments where Reflex feels great. Running across the battlefield, firing from hip, works beautifully in that you can spray fire anywhere you want, without it affecting the direction you're running in - you can't do that so easily on the HD console versions. Similarly, the process of using ironsights, the sniper scope or just lobbing a grenade is transformed completely. If the overall "aim" in using these modes is to get more precision and control, the effect is boosted immeasurably through the use of the Wii remote's infra-red pointer.

However, over and above these aspects, the awkward button placement on the controller makes things difficult for anything other than running and gunning and it's here that the control system is left wanting. Lobbing flashbangs and grenades with the plus and minus buttons involves a clumsy rearrangement of your fingers on the remote - chucking back thrown grenades is particularly time-consuming and unwieldy. Attaching the knife to the down button on the d-pad also kills off the whole notion of having an instant form of close-range kill due to the time it takes to move your fingers into position. Over and above that, there is still the notion that simulating head movement by pointing the view weapon to the far left or far right of the screen feels rather unnatural.

That said, there is something here that - when it works - feels worthwhile, and it makes you wonder just what sort of implementation we could see with the Sony motion controller. If the system's mooted PlayStation Eye head-tracking tech (coming to GT5) works well with minimal latency, this could solve the view weapon issue. That, combined with the extra precision from the Sony controller, could make all the difference. A patched up, motion-controlled Modern Warfare on PS3? That's something I'd love to see.

Look out for our review of Call of Duty: Modern Warfare Reflex, with more on the online modes in particular, soon.

We've been planning to put this together for ages, but the recent release of Grand Theft Auto IV: Episodes from Liberty City (reviewed today) proved to be as good an excuse as any to finally put the effort in - not just because the additional content is so cool, but also to pay tribute to what remains the finest, most expertly realised open-world environment ever to hit consoles.

Just sitting back and watching the world go by in GTAIV is an experience in itself, and the range of behaviours attributed to the gameworld's pedestrians is absolutely phenomenal. They drink coffee on the way to work, they sit and read on the benches, they warm up before going jogging, they put their brollies or run for cover when the rain comes, they get immensely annoyed if a car runs into them.

Some of them smoke, some of them don't... They'll even pick fights with one another on the odd occasion. Combine this with the realistic movement of each and every vehicle, along with the uncanny lighting system and the superb realisation of multiple weather types and you can't help but appreciate the sheer technical accomplishment.

The player's focus is of course on the game itself, and the cityscape that Rockstar North has created is taken somewhat for granted bearing in mind just how far ahead of the competition this technology is. Creating and coding the make-up of Liberty City must have been a mammoth undertaking, but in the midst of the involving gameplay it's just background detail as Niko's story unfolds - or Johnny Klebitz's, or Luis Lopez'.

The PS3 version of GTAIV was used for this presentation for a couple of reasons: firstly, we had a save-game with the camera-phone! Secondly, the lower resolution and omission of odd texture-dithering compared to the Xbox 360 game helps video compression immensely. Be sure to check out the HD version for all the intricate details.View this video in HD

There are complaints that the game feels laggy and that the frame-rate is somewhat variable and often disappointing, and it's difficult to argue with that, especially when other open-world games sustain their frame-rates more convincingly. However, it's important to remember that in addition to everything we've just described, Liberty City is a creation unlike any other.

Games like Prototype or Crackdown can stream and decompress data relatively easily in comparison thanks to the multitude of repeated graphical assets, shared textures and more basic geometry. Not only that, but the developers can shape the environment as they please to match the limits of their technology. GTAIV on the other hand is attempting a full-on recreation of New York City, with all the challenges that represents.

Speaking of challenges, putting this video presentation together wasn't exactly a walk in the park. First of all, we needed a technique to sustain a first-person viewpoint without the game shifting into spectator mode. This was solved by using the camera-phone you get later on in the story. Next up, the capturing. GTAIV's day-night cycle lasts for about 50 minutes of real time and we captured around 45 different clips at one frame-per-second over the course of 10 days, not all of which made the cut.

While our previous time-lapse videos have simply faded between day-night cycles, for this one we went a little further. Thanks to the on-screen timer on Niko's mobile, we were able to edit together 15 days of game time - complete and unabridged - encompassing 35 different locations from Liberty City.

It's our tribute to a superb technical achievement that no other open-world game has yet to match, and we can but wonder what the team at Rockstar has planned for the inevitable next GTA offering. With the technology in the bag and still essentially unrivalled, will the developers shift the focus to content creation? Or can we expect an even more advanced version of this astonishing engine?

At around 2.6GB in size, the God of War III E3 demo is one of the meatiest sampler downloads we've yet experienced, but the lengthy wait is worth it bearing in mind the length and breadth of the content on offer. Indeed, there's so much to show that we've cut the video analysis into two parts with only minor edits.

The first effectively covers what was shown in the Sony press conference at E3, the second shows what comes afterwards, and it's here that the demo really comes into its own - becoming that much more of a technical showcase. If you've already eagerly consumed the E3 press materials, skipping ahead to the second vid is probably worthwile.

This video essentially replicates the God of War III playthrough as seen in Sony's E3 conference. Skip ahead if you want to see new stuff, hit the play button to check out the frame-rate analysis.View this video in HD

So, part one then. Our initial technical analysis of the E3 demo remains relevant (it is the same code after all), but what is noteworthy is just how clean God of War III is. It's not really so self-evident on compressed internet video assets.

The texture work is of an outstanding quality throughout, special effects are used in a subtle and reserved manner, lighting is exemplary. The code appears to be using 2x multisamping anti-aliasing, but similar to Killzone 2, the choice of colour palette helps in adding to the edge-smoothing effect.

The second part of the E3 demo playthrough. This is where things begin to get seriously impressive. March 2010 can't come quick enough.View this video in HD

Into part two and the Sony Santa Monica studio's range of effects comes to the fore a touch more prominently. The per-pixel lighting in combination with the high-quality texture work produces some uncanny effects (for example, on the marble). The depth-of-field effect isn't "in your face", it just works and looks superb. The texture filtering employed is of an excellent quality. The only downers are a few low-poly edges and the reduced alpha buffers, but the impact on overall image quality isn't really a problem at all.

So, the frame-rate. Yes, it's E3 code. Yes, there's obviously a better-than-usual chance that the final code will improve over what we see in the demo. However, the performance level in the sampler is intriguing. This rendition of God of War III is v-synced (so no tearing), and based on our playthrough across the two videos, we have an average of 36.81FPS, a low of 24FPS and a high of 56FPS.

Average frame rates aren't usually that much of a useful stat, but here it's a pretty decent indication of overall performance throughout - which is curious. Locking the game at 30FPS would have produced a more visually consistent look, as well as a more predictable, "reliable" feel from the controls - plus less judder on-screen in panning shots and the like.

Overall though, small quibbles aside, this demo is great stuff, and the timing of its release just weeks before the US demo is released as part of the God of War Collection is curious. Will this self-same E3 demo be the same sampler included as a PSN redeem code within the package? Has Europe actually had the "exclusive" here?