Tech Analysis: Dante's Inferno

Dante’s Inferno may be a blatant God Of War rip off, but it is also one of the best examples of platform parity across PS3 and Xbox 360. It does so not by playing to the strengths of each machine, but by simply having an engine which barely taxes either system, making some concessions to alleviate the issue of PS3 having a lack of available bandwidth, and 360’s need to fit the framebuffer into it’s 10mb worth of EDRAM.

You could almost say that Visceral Games effort is an almost exemplary example on how to get a game running and looking identical on both platforms, or rather almost 100% identical. The only exception we noticed to that being a slight blurring of the image on 360, but more on that later.



Dante’s Inferno runs at a flawless 60fps on both PS3 and 360, with no noticeable sign of screen tearing or framerate drops, which in it self is quite impressive for a multi-platform title. However it does this through using only a limited number of memory and shader intensive effects. So what we have here is mostly flat looking textures, with bump mapping reserved for the characters and only certain parts of the environment.

The game is rendered on both PS3 and 360 in 720p (1280x720) with no anti-aliasing of any kind. This allows the framebuffer to fit into the 10mb EDRAN found on 360’s GPU, whilst making the conversion to PS3 much easier as it doesn’t put a stain on the bandwidth. The resolution of transparencies and particle effects usually lowered on PS3 due to the lack of available bandwidth has been compromised on both versions. So instead of 360 having the usual advantage when it comes to displaying loads of multi-layered effects, it’s merely equal across the board. Again this basically allows the smooth running of the game on both platforms whilst keeping the actual look identical; certainly, it’s how Visceral have achieved the constant 60fps on display.

Anisotropic filtering and texture detail is like for like across both versions, demonstrating clean and clear characters and vistas, though the overall sense of scale is rather small, and the detail itself is somewhat simple when compared to the likes of Bayonetta or Devil May Cry. Serviceable is how I think you could best describe the overall look and technical application.



Now earlier we mentioned that both PS3 and 360 games were almost identical, except for a slight blurring on the 360 version. This blur whilst being hardly visible during fast moving scenes can be clearly seen in the still screenshots above, and during more sedate moments of gameplay. It seems that there is a horizontal 1-pixel wide blur on all edges, with no apparent reason as to why. It could be that the developers still wanted some sort of AA solution, but seeing as 720p 2xAA may not have fit into the EDRAM, they thought a simple blur approach would suffice. It’s perhaps the only blemish on what can be considered one of the best multi-platform conversion examples available on both consoles.

Overall, Visceral Games have shown just how to successfully accomplish a good multi-platform conversion without sacrificing too much from each version along the way. Sure they could have played up to the 360’s strengths and added higher-resolution effects and more particles, or had extra HDR lighting on the PS3 game, but it would have taken longer to develop and required more optimising for both versions. This is a problem most would rather avoid, so it’s easier to go down the safe route, and keep your development budget under control and get good results, rather than having it spiral out with two different versions, each having their own tweaks, and neither achieving parity.

Date’s Inferno shows you don’t need to achieve a massive technical accomplishment when creating a game, but rather just a well thought out approach and a solid underlying engine, which can perform on both systems without needing to radically tailor features to each one.

In this respect Visceral have been successful, and I imagine that more developers will go down the same path seeing as it can work so well.

Editorial: Can Sega Make Sonic 4 Succeed?

Yesterday Sega unveiled Sonic The Hedgehog 4 to the world, and in flurry of excitement and burning anticipation we brought that news to you at IQGamer. Today the dust has settled and a dead calmness has set in, bringing with it feelings of anxiety and doubt as to whether what we saw yesterday could really live up to its promise. The legacy sewn by those faithful four original Megadrive games is not something easily replicated, even for Sega who produced the graphically impressive but rather flawed Sonic CD some 17 years ago.

Given Sega’s track record these last few years, filled with failed 3D attempts to capture that 16bit essence, a disastrous reboot in the form of a next-gen Sonic Adventure sequel, along with a handful of outsourced 2D handheld instalments which seem to completely forget just what Sonic games were all about, my cause for concern is far from being misplaced.



Having this all-new 2D Sonic game as a sequel to Sonic and Sonic & Knuckles is a tall order to fill. Creating a canonical successor in a similar style and with respect for the source material is no easy task, especially some 16 years on. Only a few developers such as Capcom with Street Fighter IV, Mega Man 9 and Bionic Commando Rearmed have truly succeeded in doing this.

Sega looks like it’s trying its best with what they have, and in an interview with GameSpot, Ken Balough seemed to be addressing most of my concerns, along with other die-hard fans. Essentially he told GS that the team was going back to the MD games, and using them as the template for Sonic 4, however the style and tone of the game would be as if the series was created today rather than 16 years ago.



In the trailer released the visual look of Sonic 4 was clearly a natural follow on from the likes of Sonic 2, with a hint of Sonic 3 and demonstrating a modern twist on the proceedings. However there are certain things which don’t look quite right. Sonic’s character model for starters looks a bit iffy, almost like a throwback to the design used in Sonic and the Secret Rings, and currently suffers from some uncomfortable running animations. The stage shown in the trailer also looked a little too much like it was from a well made mugen game, and lacked the style you'd expect from a true sequel to Sonic and Knuckles, whist also feeling slightly unfinished from a graphical perspective. However the inclusion of parallax scrolling in the background brought a twinkle to my eye, and it was rather impressive seeing the effect updated using todays hardware. From a stylistic point of view though, Surely Sega should be looking at replicating a new take on the style featured in Knuckles Chaotix, or at the very least Sonic 3, especially as Sonic 4 is supposed to be a proper game in the series?



Maybe this will change as development progresses, though seeing as the release is only a few months away I have a sneaking suspicion that it probably won’t. Still, at least we have a style that fits in with the original Sonic games, with an updated look for the current generation. It could be much worse.



Gameplay wise it was most certainly impossible to gauge just how Sonic 4 will pan out. With only a couple of seconds of footage showing very little other than Sonic briefly jumping on a few enemies, and speeding through a corkscrew platform before reaching the end of the stage. Although comments from Sega’s Ken Balough can allow us to at least assume that they have forgotten about Sonic being about speed, and will instead be focusing on having cleaverer level design based around building up momentum; finding that ‘perfect path’ through the level, and effectively achieving max speed via skill instead of by simply holding down the d-pad in the forwards direction.

This is what needs to happen, as this insistence on speed is what killed the GBA games, and made Sonic Rush nigh on unplayable for anyone used to the MD games of old.

Sega should also create new enemies in the style of the old ones, rather than simply rehashing and updating existing badniks for the 21st century. Old bosses should be given new tweaks, and mixed up with some innovative new ideas all in keeping with what the classic franchise is known for. Maybe they should look towards both Sonic CD and Knuckles Chaoix for inspiration, as both these games had more elaborate set pieces and boss designs than their MD counterparts. In essence creating a really authentic follow up in the form of Sonic 4.



Musically rather than just also reworking old themes as with the graphics, Sega should ideally be looking to create original new music for both the title theme and the tunes to be used in the various stages themselves. A new theme tune, if they come up with one, should be in keeping with the style and direction the series was heading in with Sonic & Knuckles, whilst taking care not to parody the series iconic sounds, and simply build on the foundations laid down all those years ago.

Ultimately this latest instalment cannot be just a cheap fan service attempt, certainly not if Sega expect it to be taken seriously as a proper Sonic 4, although that would still make for a potentially lovely game, it wouldn’t do the franchise any real justice. And that is what they need to show, if they are to redeem the brand and successfully move it forward looking to the future.

Anything that fans do complain about, or which they feel maybe isn’t quite right, I do expect Sega to listen to and correct in further episodes. I also expect they’ll expand upon the things that worked in the first episode, the style and ideas explored, whilst evolving them forward in preparation for potentially Sonic 5. That is, of course if they manage to hit the ground running with part 1 of Sonic 4.

Either way, Sonic 4 represents a trial and proving ground to see if a revival can be done, and if Sega still has what it takes to make this happen.

For the sake of the character and the franchise I really do hope so.

Sega Announces Sonic The Hedgehog 4

Today SEGA finally revealed just what Project Needlemouse is all about, and unlike what you may have thought; it’s not just a return to the character’s 16bit roots, but also a fully-fledged sequel to Sonic 3 & Knuckles. Say hello to Sonic The Hedgehog 4!

Sonic 4 is to be released as an episodic series of download titles making its way to all three major platforms (PS3, 360 and Wii), and will be a proper 2D game created using 3D graphics and visual effects. Both 360 and PS3 versions will run natively in 1080p whilst the Wii version gets 480p instead. Motion controls are being added to both the Wii and PS3 versions, though the PS3 game will only have mild Sixaxis usage.

SEGA also plans to rework some of the music from the classic 16bit titles, along with creating new tunes in a similar style for Sonic 4. We also assume all the classic sound effects will be brought back or updated in the same way too.

With regards to the gameplay, Sonic 4 will feature only Sonic as a playable character, and will keep his entire repertoire of moves from the first four Megadrive games, whilst including the homing attack from the Sonic Adventure series.

In an interview with GameSpot Sega associate brand manager Ken Balough stated: “We're going to deliver a Genesis-era Sonic game as if it were created today that goes to the core of what classic Sonic fans desire.”

Certainly hearing that, and seeing the short but sweet trailer, gives us hope that this really will be Sonic’s shining return to form.

Sonic 4 Episode 1 is due out for download sometime this summer.

Iwata Clarifies Wii HD, DS2 Rumours

Late last year various rumours started to materialise about how Nintendo might launch an enhanced Wii HD console shortly in order to both appease hardcore gamers, and to delayed the release of a true successor to the current system.

Today at an investors meeting in Japan, Satoru Iwata denied that he ever stated that there even was a HD version of the Wii in the pipeline. He continued by adding “If I were asked if a Wii that's been made compatible with HD -- just high resolution compatibility -- would be bought by players throughout the world, I'd of course say, 'Do you think it will sell with just that? It needs something new.”

In addition, rumblings about the successor to the Nintendo DS was also rife after Asahi Shimbun Daily reported that the next iteration of the DS would feature High-resolution graphics and motion sensing capabilities, along with at least two touch screens. However it appears that Iwata in the interview with them didn’t actually state this either. When asked the question he replied: “These are of course thought to be required. However, would it sell with just this?” however, the paper simply removed the second half of his response, according to Iwata.



Now whilst we probably can shoot down any rumours of a Wii HD, the news of a high resolution DS successor does make sense. Back in November last year, brightsideofnews.com reported that nVidia had been given the contract to work on Nintendo’s next hand-held system, specifically using their Tegra chipset. Though unconfirmed by Nintendo, and the fact that the source has been a little unreliable in the past, it does make sense that Nintendo would partner with someone like nVida, Ati or ImgTech this early along.

If a version of Tegra (a CPU/GPU on single chip) is used, it’s likely to be one of their med-to-high end varieties, and would produce visuals similar to the original Xbox or GamCube, and supports shader model 3.

As the timeframe for a new Nintendo handheld approaches, we’re sure to find out more sooner rather than later.

GT5 Possibly Delayed Again?

After $60 million and five years of development work, you’d think that Kazunori Yamauchi’s Pièce de résistance would be ready for release fairly shortly. Not so according to James Armstrong, an executive from SCEE’s Portuguese division. In an interview with Canarias Al Dia, a Spanish site, he stated that he thinks GT5 will arrive “this fall, before Christmas", before also saying that it’s "not yet decided". Right?

So when is Gran Turismo 5 likely to come out? Well, your guess is as good as ours, considering the gap between GT4 and GT5 is two years longer than the one between GT3 and GT4 on the PS2, and it’s not even out yet. It took only three years for GT4 to arrive after part 3 if you’re wondering.

Our best guess is that GT5 will arrive at the earliest, sometime this summer in Japan, with September being the very latest it could hit. Between September and Christmas sounds likely for Europe and North America.

Naturally, none of this has been confirmed by Polyphony Digital, though we wouldn’t be surprised if this there isn’t a smidgen of truth to be found.

The magical roundabout continues…

Halo Reach Engine: Early Tech Analysis

A few days ago Bungie released the first batch of in-game screenshots from their newest Halo title. Interestingly these appeared to be in full 1080p (1920x1080) resolution, and didn’t seem to be upscaled. In addition, the screens at first glance also showed off a few improvements to the Halo engine, such as possible AA (anti-aliasing) and for the first time, AF (anisotropic filtering).

Whilst it’s not uncommon for companies to release higher resolution rendered screens to the press with higher IQ than the final game, and Bungie did the same thing with Halo 3 (remember those 720p native screens?), the first official sighting of Reach’s gameplay looks to hold some truths in regards to the engine used in the final game.



The above screen was posted on Bungie.net and is obviously in 1080p. However there appears to be a huge level of AA on display, far more than the 4xMSAA used in certain 360 titles such as Dirt 2 and Race Driver Grid. This simple excess of AA can be explained away with the use of supersampling, a simple process in which an image is rendered at a much higher resolution, and then downsampled to form a cleaner smoother final image. This is basically a form of AA which requires a huge amount of processing power, due to having to render at a higher resolution than the one you want to display at, but is also very effective at eliminating jagged edges across the whole scene.

It seems that this shot was originally rendered at 2560x1440 (thanks to AlStrong for the pixel counting) with no AA, and then downsampled giving the smooth 1080p image. This is certainly something that won’t be seen in the final game. Instead we expect Reach to render either in full 720p, or using the same dual 1152x640 buffers found in the previous two games. Currently we have no idea on whether any form of AA will make it into the game, that all depends on if they’re using the same HDR lighting method from before, or if they have adopted the use of tiling in order to fit a 2xMSAA frame buffer into the 360’s 10mb edram.

Either way, in order to render in 720p or using the 1152x640 dual buffer method with MSAA, they will need to title. Especially as they claim to be pushing more enemies and vehicles around on screen than before at any given time.

However whilst we can’t judge what AA will be used (if any), or at what resolution Reach will be rendered at, it is possible to see at least 4x AF + maybe trilinear at work, or just AF with a bias towards certain textures. If you look closely at the ground you’ll see it blur much further into the distance than with both Halo 3 and ODST.

Texture detail in itself, has also been improved, with greater levels of normal mapping and improved shaders on many surfaces. Combined with the use of AF, the numerous bumps and curves of the cliff faces and surrounding scenery are much more apparent than before.

You can also find small elements of SSAO (Screen-Space Ambient Occlusion), particularly on some of the environment details, although rather subtle. It doesn’t appear to be present in the entire scene, with only a few points of visibility in other areas. A clear example can be found in the screenshot below, look at the metal shelving in the bottom right. The shading there isn’t quite correct, a usual side effect of using SSAO.



There’s no sign of the improved lighting model in the first-person screenshot either, though it’s very hard to tell, especially without seeing the same area, but from different angles as to analyse the shadowing and lighting. At present it looks very similar to ODST, and there’s no sign of the multiple independent light sources that Bungie stated is going into the game. Though those have been witnessed in other screenshots released, showing what looks to be from in-game engine cut scenes. (see above)

Arguably the gameplay build we’re seeing here is very early, and is almost certainly likely to see a noticeable improvement by the time the beta launches. Between now and then, we still have many questions that need answering. Will Bungie still be keeping the benchmark HDR of the last two games, in addition to the 30 or so real-time light sources being added? Are we going to be seeing AF plus a proper 1280x720 frame buffer. Or will the HDR be dropped for an easier 1152x640 with 2xMSAA and AF approach using titling?

We just don’t know yet, and I guess the beta is when we will find out about some of this stuff. The rest will no doubt have to wait for a more complete single-player build of the campaign, in which nothing will be paired down to deal with lag, split-screen and other concerns, which almost certainly impact on graphics.

Anyway, this small insight into Halo Reach’s graphics tech is all we have for now. But stay with as, as we’ll be following any developments very closely.