Let’s Talk About My Build

Here’s a sad fact: until just a couple months ago, my daily driver was an AMD Phenom II X6 1055T. It had served me well since 2011 and miraculously held its own up until its retirement. Honestly the chip still works mostly fine, even if its missing some core code libraries that prevent it from running a fully capable gaming PC in 2020, not to mention DDR4 and PCI Express Gen 3 compatibility. So it was time for an upgrade, and after watching many tech Youtube channels for the last few years, and practicing with a lovely low-end puppy I built for my mom for Christmas, it was time to put my knowledge to the test and build something really special that fit my needs.

The above Linus Tech Tips video asked a really important question: what are developers’ needs? I’m an indie developer, so theoretically I don’t need anything beyond a midrange PC, probably something around the $800 mark. But I figured, if my Phenom could last me over a decade with some levels of discomfort toward the end, then something with some longevity could last me a decade very happily. Considering I don’t just do game development, but also video editing, graphic design, gaming, and I’m always finding new hobbies, I wanted a PC that could hold its own no matter what I threw at it.

With that in mind, let’s talk my build.

Chassis – Fractal Design Meshify C

This case interested me for one reason in particular: filters. My old case, an Antec 900 from 2007, was wide open to the elements, and I work out of my bedroom still. That means one thing: dust. Lots of dust. Dust from laundry. Dust from my sheets. Dust from me sitting next to the computer as my skin slowly dies a sad death. The Meshify was the only case I found that has front, top, and bottom filters, meaning in two months I haven’t had to clean the thing yet, and this weekend all I have to do is take a quick dust cloth to the power supply guard and wipe down the filters. On the Antec, I’d have had to schedule time for a deep clean by this point. This case has been great. My one gripe is the cable management space in the back is a little narrow, and could use an extra quarter inch, but it’s much better than the 900, which didn’t have cable management space AT ALL. Honestly, Linus put it best.

CPU – AMD Ryzen 9 3900X

With CPU’s, I’m a big believer in the Price:Performance ratio. Yeah the 3900X is NOT cheap, but it’s about half the price of its 3950X big brother (when on sale), and monumentally cheaper than Threadrippers. It also beats out second generation Threadripper chips in processor benchmarks, and X570 motherboards are going to be compatible with newer chips for at least a few more years, meaning my upgrade path for the foreseeable future is still wide open if I need it to be. 12 cores and 24 threads of pure Team Red power might be overkill for what I do, but I did not want to end up in a Phenom situation again. I needed a chip that could handle anything I threw at it and this one has NOT disappointed.

Motherboard – ASUS TUF Gaming X570-Plus (WiFi)

I’ve been pairing up my AMD chips with ASUS motherboards since the Windows 95 days, and I wasn’t going to stop now. Going X570 means PCI Express 4 support, and great Ryzen compatibility across the board. FWIW, I’m not using the WiFi, but it’s neat to have the option.

GPU – eVGA GTX 1070 Superclocked

Believe it or not, this is one of the few holdover pieces from my old PC. Miraculously, it worked fine with the Phenom, and it works even better with the Ryzen. Yes it’s four years old now, but the GPU market has slowed down, meaning it’s only one generation behind. It’s powerful enough to run any game on High to Ultra (depending on the game) and works great for video editing, streaming, or nearly anything else, short of ray tracing.

RAM – 32gb GSkill Ripjaws 3200mhz

Again, this was a Price:Performance choice. I wanted to push as much extra speed out of my Ryzen as possible, and with AMD infinity fabric being timed to RAM speeds, I wanted to get a good amount of memory at a fast speed and not break the bank. This particular set goes on sale for around $100 every other month or so, and I still have 2 slots leftover, so if I ever want to upgrade to 64gb, it won’t be impossible.

Sound – Creative Soundblaster Audigy 5/Rx

I can hear you right now. “Wait. Sound card? What?” Yes. I’m a sucker for 3D audio and I use the Soundblaster for my headphones to get just that. On my old PC, I mainly had this installed because the sound died on my motherboard years ago. On the Ryzen, I use onboard audio for my speakers, and the Soundblaster for my headphones. Very optional but a nice addition if you’re wary of using the front audio ports.

Cooling – Noctua Everything

The Antec 900 had many flaws, including the lack of cable management, the terrible hard drive cage placement, and the dust. But oh lord, the fans were bad. The way they were set up at the front of the hard drive cages meant they weren’t replaceable, at least not without a lot of drama. I don’t think the top fan was replaceable at all. This was a big factor in me picking the Meshify C as I needed to be able to replace the fans if they ever died, and needed them to be easily removable for cleaning, fixing, etc.

My 3 primary fans (2 front, 1 top) are Noctua NF-P14s redux-1200‘s. These 140mm beasts make zero sound, do a great job at keeping things cool, and keep the grayscale color scheme. All 3 are equipped with black anti-vibration pads. The front two are also held up with anti-vibration silicone mounts, but the top two are screwed in normally just due to space. The back is a standard brown Noctua NF-F12 I had lying around, also held up with anti-vibration mounts. The fun part is each fan is also hooked up to Noctua Low Noise Adapters, so things only ramp up when it gets really hot, and the rest of the time the machine is whisper quiet.

For CPU cooling, I went with Noctua’s NH-D15 chromax.black absolute CHONK of a cooler. I really didn’t want to take any chances with the 3900X, even if it’s a pain in the ass to get installed. I chose to not install the optional back fan on it due mainly to space, and the rear case fan is very close to it anyway, so the combo of middle and rear fan takes care of the heat wonderfully.

Power Supply – Corsair CX750M

750 watts so I can have more than enough leniency? Check. Modular? Check. Inexpensive AND trusted brand? Check. PC Part Picker puts this machine at 454 watts. The general rule is to multiply expected wattage by 1.5 and that’s the power supply to get. In my case, I wanted to go modular, and this was the best choice on the market in that 700ish watt range. The extra 50 also gives me a little leeway for the future in case of graphics card upgrade or anything.

Storage

I have a bunch of storage. My boot (and currently software) drive is a 1Tb Samsung 860 Evo, actually reused from my Phenom PC with a miraculously successful Windows 10 transfer. I then have about 6Tb of mechanical storage for work, music, photos, etc. and a 10Tb external backup drive. I’m likely going to invest in some more SSD’s soon, including an M.2 drive for my games. I can technically have 2 but I’d have to take off the chonk of a CPU cooler to reach the second one and I’m in no rush to do that.

Full Specs List

https://pcpartpicker.com/list/3ct4L2

A quick note about the above link: PC Part Picker lists the value of my PC at over $2000 😲. I did not spend NEARLY this much on it. Yes a good number of components are reused from my old PC, but the big killer here is the graphics card. Since the GTX 1070 Superclocked is no longer available, it’s listed as $625. DO NOT PAY THIS MUCH FOR A 1070!!! You can get an RTX 2070 Super for $5-$15 LESS. So for the love of God, save your pennies and boost your power.

If you’re looking to build a more moderate-powered PC, swap the Ryzen 3900X for a 3600X, saving $200. Go for a slimmer profile CPU heatsink, probably saving more money in the process. Do not skimp on the RAM, the 32Gb of GSkill at 3200mhz is perfect. Pick up a GTX 1660 Super to match the graphical horsepower of the 1070 AND save $400. Forego the sound card and mechanical drive. At that point you’re looking at more like a $1200 PC which will still take you very far.

If you want a low-power PC that still bounces like a puppy, well, that’s another post for another time.

– Danny

Let’s Talk About E3

I still remember 2007, that dreaded year when E3 was gutted and shoved into the Barker Hangar in Santa Monica. Nobody showed up. Nobody cared. This wasn’t the E3 we knew, or wanted, no matter if you were a developer, a journalist, or just a gamer. It sucked. It felt as if the life was being sucked from the one event that DEFINED gaming. Being in college, studying game development at the time, my friends and I couldn’t help but feel discouraged by the event. PAX was barely a thing and PAX East was still 2-3 years away from happening. GDC was industry-only. When it came to gaming conventions in America, E3 was the place to be, and it was about to die. Heck, a lot of photos of the convention have been scrubbed from Google Image Search, as far as I can tell. In short, we don’t talk about the E3 Media & Business Summit. It never happened.

Fast forward 13 years. E3 made a powerful comeback. We have announcements of new consoles (XB1, PS4, WiiU AND Switch), incredible game announcements (see: God of War), and even breathtaking memes. E3’s resurgence over the past decade has been great. The show got over some major stumbles and returned to being the biggest holiday in gaming.

Then E3 2019 happened. Media attendees were inadvertently doxxed and their contact info leaked out all over the internet. Sony was nowhere to be found. Nintendo was comfortable with their digital-only presence. Now E3 2020 is on the way, the list of attending exhibitors has leaked, it’s known that the show will be placing celebrities and streamers in the crowds to try to hype them up in the lines, there will be a Lakers exhibition game, and Geoff Keighley, creator of the Game Awards and a longtime E3 attendee, has publicly announced he will not be going. Normally one person not attending a conference isn’t a big deal, but Geoff puts together behind-the-scenes interviews and information that no other media personality does at the show. Will someone pick up the slack? Maybe, but that is yet to be seen.

Geoff states that E3 needs to be more “digital, global and inclusive.” I agree, but the same could easily be said of Gamescom and Tokyo Game Show, which cater more toward European and Asian developers respectively. While I think it would be good for E3 to lead the charge, they have chosen their direction. The other big conferences are sticking with what has worked traditionally, while E3 is going to focus much more on streaming, celebritydom, and basketball. In essence, E3 is going to be about LA, and not about games. Games just happen to be what the show has been about, but I bet we see that change going forward.

The point is, E3 – sorry, LACon – is starting to drift back to those dark 2007 days, but possibly in an even worse manner. At least in 2007, the focus was still games. Now the focus is Los Angeles, Hollywood, and superstardom. Other than developers in Los Angeles – and yes, there are plenty – I don’t think many of my fellow game developers would be enthusiastic about attending an event like that. When we founded BostonFIG in 2012, we wanted to cater to Boston and New England-based developers, specifically because the only other local option – PAX East – was filled up with every other developer from around the world, and our burgeoning community didn’t feel like it had the home there it should have. With more conferences springing up, E3 has competition. With hundreds of new games released every day, conferences in general being a pain for everyone to attend, and the media looking to go elsewhere just for the sake of keeping their addresses off the internet, developers and publishers are simply going to walk away and never look back. We don’t want LACon. We want E3.

Published
Categorized as Blog

Let’s Talk About the Game Awards

The Game Awards, hosted by Geoff Keighley, were last night and I have some thoughts. I’m not normally one to put my thoughts down in writing, but I can see the direction the media is pushing the public image of the industry and I think it’s worth keeping an eye on if you’re a developer.

I noticed the trend a few years ago, when big companies started to sponsor streamers and eSports players. Thousands of dollars just to run their logo across the screen or wear their t-shirt on camera. Meanwhile, developers still needed to go through the process of finding a publisher or investor in order to fund their work, and 99/100 times would get rejected. The big money was in streaming the games we poured months or even years into, just to get one video out of the popular crowd if we were lucky.

Last night’s game awards sealed the deal. 29 awards in total, and 6 of those were devoted to people and THINGS in streaming and esports, including:

  • Content Creator of the Year
  • eSports Coach
  • eSports Event
  • eSports Host
  • eSports Player
  • eSports Team

I get it. eSports are huge. Gigantic even. But for an event that supposedly celebrates the art and innovation of games from the last year, where is the Programmer of the Year? 3D Modeler? Composer? Concept artist? If you’re going to have an eSports Team award, why not Studio? eSports Event but not Games Expo? eSports Host but not Community Manager? eSports Coach but not Studio Head?

As Donald Mustard from Epic Games pointed out last night, Fortnite has over a thousand staff working on various aspects of the game day in and day out, from development to tech and community support. The game overall won a few awards, and the staff is surely happy about that, but at the end of the day a team is still comprised of individuals. How are those individuals’ efforts less important to the game industry than that of a single gamer who just happens to be very good at the games those developers created?

While I understand how quickly eSports is growing and how important it is to recognize the achievements and contributions of individuals in that sector, I find it bewildering that individuals who create unbelievable works of art every day are left by the wayside for what has become a very important, annual event in their own line of work. The Oscars have awards for Costume Design, Editing, Hair & Makeup, Score, Production Design, Sound Editing, Sound Mixing, Visual Effects, and Writing. There is no separation by genre, and they even have smaller awards shows dedicated to Science/Technology and Student Films to praise other areas of filmmaking. If The Game Awards wants to be the Oscars of gaming, they absolutely need to follow suit. Start praising the parts and people of game development, stop the antiquated “Best RPG” and “Best Racing Games” separations. It’s arbitrary, and so many games are blending genres now that it’s insulting to compartmentalize them.

If I were in Keighley’s position, I would split the Game Awards into 2 or even 3 events: AAA, Indie/Modding, and eSports. Start praising the achievements of teams and individuals in actual game development rather than general genre awards and neglecting the hard work of artists worldwide in favor of internet celebrities.

It’s clear to me that eSports are starting to creep their way toward taking over the awards, and while they are important, the games they play are moreso. I think it would be a good idea to devote shorter shows to each sector before even Game of the Year gets left by the wayside.

Mondrian Dynamics Part 2 – Risk Reward

“Difficulty” is a tricky thing to describe when it comes to videogames. Plenty of games have one base-level of difficulty throughout, and it can’t be adjusted, so you better get good. Other games follow the Easy/Medium/Hard model (or sometimes “I’m too Young to Die” through “Ultra Nightmare!”). Recently we’ve seen the creation of assist modes such as in Celeste or Mega Man X Legacy Collection to remove nearly all punishment for failure, and make playing the game easier for newcomers. No matter what though, Difficulty has traditionally been a static variable for one or two mechanics, such as Enemy or Player Health, or the availability/strength of items. In Mondrian, Difficulty is dynamic, and nearly every system in the game adjusts depending on how tough you are.

But seriously, how tough are ya?

When beginning a new game, players can select from 4 difficulties: Relaxing, Easy, Medium, or Hard. Increasing the Difficulty has the immediate, visible effect of opening up progressively tougher variants of each level, and in fact, in Mondrian Maker, you can iterate the Difficulty in order to change your level’s Block Types and Modifiers, with up to 4 different variations per level (Obstacles are not allowed on Relaxing, but Modifiers are). The game will always load a variant less than or equal to the current difficulty; for instance, if you’re playing on Relaxing, you will only see Relaxing variants, while if you’re playing on Easy, you’ll see both Relaxing AND Easy variants. Essentially, the player is rewarded with more content for taking the risk of playing at a higher difficulty.

Difficulty also has some effects on the levels themselves. Gray Blocks take multiple hits to destroy, and the higher the difficulty, the higher their health. Splitter Blocks break into multiple, Regular Blocks when hit, and the higher the difficulty, the more Splits they create. Modifiers are not directly linked to the Difficulty value, but you can adjust their timers manually per Difficulty to create tougher effects, suited to your personal tastes.

There are, of course, other systems at play related to Difficulty, such as how likely a Gem is to spawn (lower chance at higher Difficulties) and the maximum value of those Gems (higher value at higher Difficulties); how many blocks it takes to make the Speedball go faster; how much the Breakerball heals when hit with the Paddle or a Gem is collected; how much the Stretch paddle scales when a Gem is collected and how quickly it shrinks the rest of the time; or how much action is allowed on screen before the ball will gently rest in place, giving you a chance to take in the particles and collect Gems. Essentially, we are making sure as many game systems as possible are tied to the Difficulty value, where appropriate, in order to scale the entire game and not just one or two systems.

Of course, this is just the Difficulty value. The Game Speed value, selectable under Gameplay Options instead of at the start of the game, can also be seen as somewhat of a Difficulty value, though it has more of an impact on the feel of the game than on the mechanics. Whether the game is played on Slow, Middle, or Fast has no effect on, say, the chance of Gems spawning, just how quickly they’ll try to fly out of the field. Game Speed is tied to anything with movement, including Paddle Spin, Stretch Paddle scaling, Ball speed, the groove rate of blocks, and even the soundtrack! That’s right, if you’ve never changed the game speed in Mondrian, you may want to just to hear totally different tunes.

By using Difficulty and Game Speed in a dynamic fashion, and tying them to every game system instead of just one or two values, Mondrian is able to scale pretty nicely across a broad spectrum of skill levels. It’s still a tough game, and we’ve even heard some complaints about the toughness, so we’re in the process of adjusting some math. Our goal here is to not make the game too easy – we still want it to be a challenge! – but to smooth out some curves of unexpected behavior, speed up the end of levels, and make moment-to-moment gameplay act a bit more predictably.

Since Cuphead launched in 2017, there’s been a debate about how difficult games should be. Hidetaka Miyazaki, designer of the Dark Souls series, has said that his goal is to give players a sense of accomplishment by making them overcome tremendous odds. Some designers like things that way, setting how the game plays from beginning to end. Some designers like adding in difficulty settings that change one or two variables throughout the game, while others take the halfway approach, having an “envisioned” mode but offering some assist options if players need them. I like to think of difficulty as a number, not an adjective, and seeing how math can shape an experience. In the case of Mondrian, I’ve always wanted players to be able to play the game how they want to play it. Want to chill out at 3am with some slow jams and pretty colors? Mondrian can do that. Want a high octane action/puzzle experience that puts Polybius to shame? Mondrian can do that too. We want Mondrian to have a little something for everyone, but with a core ruleset. No matter what speed or difficulty you’re playing on, we want you to get familiar and comfortable with those rules, but more importantly, the relationship between those rules. Much like Mondrian’s color wheel, our aim is to make the game’s dynamics work together in harmonious beauty.

Mondrian Dynamics Part 1 – Color Theory in Action

Everything is expressed through relationship. Colour can exist only through other colours, dimension through other dimensions, position through other positions that oppose them. That is why I regard relationship as the principal thing.

– Piet Mondrian

Mondrian – Plastic Reality is an orbital brickbreaking game where players explore the lives and works of five modern artists. The game is the sequel to our surprise hit Mondrian – Abstraction in Beauty from 2015, and one of the goals with this one was to take a greater level of control with regards to the game’s dynamic generation. Instead of just randomizing everything, we wanted to create a harmony in the randomness. This started with color, and the relationship between colors in particular.

Mondrian – Plastic Reality’s Color Wheel controls the visual relationship between all colors on screen in the game.

The Color Wheel is the simplest method, in art, of defining colors, no matter how we are classifying them: warm or cool, bright or dark, bold or soft. Mondrian also has a color wheel, but defines its color through code instead of on a palette. The additive RGB color space covers 16,777,216 colors (256 red * 256 green * 256 blue). However, for the purposes of scoping, Mondrian, like a classic color wheel, has 12 colors: Red, Orange, Yellow, Chartreuse, Green, Spring, Cyan, Azure, Blue, Violet, Magenta, and Rose. When defining each individual element’s color in-game – blocks, the paddle, the background, etc. – we set each color channel – R, G, or B – as the Primary, Secondary, or Saturation channel. Effectively, the Primary and Secondary channels control the brightness and hue of the element, and the Saturation channel controls how gray it will be.

Red, Rose, Magenta Adjacent Color Scheme

A max lightness is set for the entire level (randomized between 127-255), and then each element’s Primary channel is randomized between 64 and that max lightness. The secondary channel, if needed, is then set to 1/2 the Primary. The Saturation channel is then set to a Random value between 0 and the Secondary channel’s value. The result is a wide range of colors, though more in the range of 7 million colors than 16.7 million.

Blue, Yellow, Orange, Azure Tetrad color scheme

Before colors are set, however, the game must know what colors will even be allowed. This is performed via a dice role for visuals. Color Theory comes into play with the Analogous Color Schemes, where the game can determine if colors should be Monochrome (one color), Adjacent (three colors in a row in the color wheel), Triad (three colors that form a triangle in the color wheel), or Tetrad (four colors that form a rectangle in the color wheel), and even whether or not to include Compliments (opposite color to the starting color). A few other schemes are included as well, including De Stijl (Red, Blue, Yellow, Dark Gray, Light Gray), Grayscale, and Totally Random. Using this, one randomized starting color, and 128 possible levels of brightness, the game is able to generate 18,179 possible color schemes. If you also multiply the background being able to be a full color or one-color variant, that’s over 36,000 possible looks for a level, via instant dynamic generation alone.

Recreating Johannes Vermeer’s Girl with a Pearl Earring in Mondrian Maker

However, if dynamic generation isn’t your thing, you can also paint your levels. Painting your levels disables dynamic color generation on blocks, effectively meaning those blocks in your level CANNOT be randomized and will always be generated with their painted color, no matter what. This can be used to create “pixel art” works, abstract puzzles, pseudo 3D effects, whatever your creative heart desires.

Pseudo-3D achieved via Mondrian Maker’s painting tools

Dynamic generation is a core part of Mondrian, but in order to achieve an aesthetically pleasing balance between the hand-crafted and the computer generated, it’s necessary to set limits and define relationships of color in the code. Most designers think that dynamic generation is limited to gameplay, such as random level layouts in games like Nuclear Throne or from prebuilt chunks like Spelunky, but our goal is to create dynamic systems that truly stretch the bounds of what art can be. Manual & automated creativity can work together to create beautiful imagery. Relationship is the principal thing.

Azure, Green, Orange, and Rose Tetrad color scheme

If you enjoyed this post, please consider Wishlisting Mondrian – Plastic Reality on Steam or grabbing an Early Access version of the game on itch.io or GameJolt.