Troy Davis (@GPFTroy) profile picture
Arizona
Joined August 2014
Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)Photo by Troy Davis (@GPFTroy)
I run deposit strategy at Chase. I have some insider information regarding Monday's bank outage.   Everyone else in the building had a bad Monday. I had the best day of my career. Nobody could take any money out.   Here's what happened. On Monday morning, Chase customers started reporting that they couldn't log in, move money or withdraw funds. By 2 p.m. Eastern, nearly 4,800 people had reported trouble with Chase on Downdetector. Zelle payments stalled across the country. At least 1 branch couldn't take a deposit. The AP reported that "a technical issue at JPMorgan Chase caused ripple effects throughout the financial system."   Our statement said, "Due to a technical issue, some transactions are taking longer than usual to complete but customers still have access to their accounts and funds." From my desk, that's completely accurate. The funds were in the accounts. The accounts were in the bank. Everything was accounted for. It just wasn't going anywhere.   Let me explain how a bank works. You give us your money. We pay you 0.01% for it. That's the rate on a standard Chase Savings account. On $10,000, you earn $1 a year. Then we lend your money out and invest it. In the third quarter, our net interest income was $27.0 billion. That's about $294 million a day.   On a normal day, your $10,000 earns you about a quarter of a penny. It earns us more. Monday was a normal day for us.   Here's a number you're not supposed to know. When we park cash at the Federal Reserve, the Fed pays us 3.90%. We pay you 0.01%. The Fed pays us 390 times what we pay you, on money that came from you. That arrangement didn't go down on Monday either.   The spread doesn't need the app. It doesn't need you to log in. It runs whether you can see your balance or not. Your money earned for us all Monday while you hit refresh. Our net interest income has never had an outage. I'd know. Watching it is my job.   Some people think an outage is bad for a bank. Think about what a deposit is. It's money that stays. My entire department exists to make money stay. We run promotions. We offer bonuses for direct deposit. We design apps that make moving money feel like work. On Monday a technical issue did my whole job for free, nationwide, before lunch.   I'd like to talk about the glitch. In 2024, customers found a different glitch at our ATMs. It let them deposit a fake check and withdraw the cash before the check bounced. TikTok called it the "infinite money glitch." We called it "fraud, plain and simple." Then we sued. In our lawsuits we asked for the money back, plus interest and overdraft fees.   Monday's glitch ran the other way. It kept customers from withdrawing money that was really there. That's a technical issue. I want to be clear about our policy here, because the public keeps confusing the 2. When a glitch lets money leave the bank, somebody is going to court. When a glitch keeps money in the bank, somebody is going to have to try back later.   Back in 2024 we said, "We're pursuing these cases and actively cooperating with law enforcement to make sure if someone is committing fraud against Chase and its customers, they're held accountable." People online have asked who will be held accountable for Monday. A technical issue. We looked into suing it. Technical issues don't have addresses.   There's one more difference between the 2024 glitch and Monday's. When we sued the TikTok customers, we filed in federal court in Los Angeles, Houston and Miami. Your Deposit Account Agreement says your disputes with us "will be resolved by binding arbitration" and "not through litigation in any court," except small claims. You also agreed never to join a class action. So if Monday cost you a late fee, you can't sue us with your neighbors. You can file alone, in private, in front of an arbitrator. There is an opt-out. It's in the agreement. We know how many people read the agreement.   Now, the fees. Say you scheduled a transfer to cover your rent, the transfer took longer than usual, and your account went below zero. Our Overdraft Fee is $34 per transaction, up to 3 a day, for a maximum of $102. We charge it during nightly processing. People have asked whether nightly processing was affected on Monday. I can't speak to that. I can tell you nightly processing is very important to us.   In December 2024 a federal agency finalized a rule that would have capped overdraft fees at big banks at $5. The American Bankers Association championed the repeal. In May 2025 Congress repealed the rule, and the President signed it. Our fee is still $34. In 2025 we collected about $1 billion in overdraft fees. I keep that number on a sticky note on my monitor. It's the most reliable system we have.   We also offer Overdraft Assist. You pay no fee if you're overdrawn by $50 or less at the end of the business day. On Monday the app didn't work, Zelle didn't work, and some branches didn't work. The business day ended right on time. It always does.   A man in Dallas tried to deposit a check at a branch on Monday. The transaction couldn't be completed. I understand his frustration. Normally when you deposit a check, we hold part of the money for a day or 2. On Monday we held the whole check, at the counter, with him still attached.   He told CBS the teller told the customer ahead of him that the "system was down." Those are 3 words we've spent a lot of money keeping out of our branches. The system was taking longer than usual. The teller has been scheduled for messaging training.   Some customers couldn't see their routing numbers. One told Gizmodo he could log in but couldn't find his. Without a routing number, your money can't be routed. That's the safest your money has ever been.   Last year we told CNBC that fintech middlemen like Plaid were "massively taxing" our systems with unnecessary pings. So we started charging them to access your data. On Monday nobody had to ping us. Our systems taxed themselves. We're still deciding who to invoice.   Bank of America and Citigroup didn't have an outage, the AP reported. Their customers just couldn't send money to ours. Shares of JPMorgan, Bank of America, Citigroup and Wells Fargo all fell around 2%. We broke, and our competitors' stock fell with ours. When you hold this many deposits, your technical issue becomes everyone's technical issue. I find that moving.   Zelle is owned by a company called Early Warning Services. Early Warning is owned by 7 banks, including us. On Monday Early Warning said the Zelle network "was fully operational." A Zelle representative told NBC the problems originated with Chase. A company we co-own named us in public. I've asked legal to remind them who the owners are.   The company is called Early Warning. I've always loved that name. It's reassuring, and on Monday it didn't have to warn anybody.   Zelle has a strong record. In December 2024 the CFPB sued us, Bank of America, Wells Fargo and Early Warning over Zelle. It said customers of the 3 banks had lost more than $870 million to fraud on the network since 2017. In March 2025 the CFPB dropped the case. When a scammer moves your money on Zelle, it arrives in seconds. When you try to move your own money on a Monday, it takes longer than usual. We've never confused the 2 priorities.   Chase has about 94 million customer relationships in the U.S. Nearly 4,800 people reported problems on Downdetector. That's about 0.005% of our relationships. On my dashboard, that rounds to 0. Downdetector is a website where people go to report their feelings. The deposits never filed a single report.   In the third quarter we made $15.3 billion in profit. That's about $166 million a day, including the days when everything works. Monday won't show up in our earnings report. Outages never do. That's how I know it was a good day.   The board paid Jamie $43 million for 2025, a 10.3% raise. The New York Times estimated his total 2025 haul, counting stock gains and dividends, at roughly $770 million. At our savings rate of 0.01%, you'd need to deposit $7.7 trillion to earn that much interest in a year. We don't hold that many deposits yet. I'm working on it. Monday helped.   For the record, my own money was fine. I keep it at a credit union.   At 4 p.m. my team had cake. Deposit outflows on Monday were the lowest anyone on my floor could remember. Somebody suggested we schedule a technical issue every quarter. Everybody laughed. I wrote it down.   My job is to keep your money at Chase. On Monday I didn't have to do anything. Your funds were safe. They were with us, which is where we prefer them.   Please try back later.
253
651
99
2,828
1,355,085
Laserdisc games on the #Dreamcast? @GPFTroy has indeed done what some might have thought impossible! Dragon's Lair (Original Arcade LaserDisc) game is now fully playable and here's a look! youtu.be/daDrlr5WiAE @falco_girgis @TheDCJunkyard @timeextension64 @IanMicheal10
4
8
29
1,198
It is with great pride that I can finally announce that my fast math, vector, and linear algebra library for the Sega Dreamcast, SH4ZAM, has now officially been added to the KallistiOS SDK within kos-ports, ready for use within any KOS-based project looking to make perf gainz! 💪 Featuring dual modern C23 and C++23-based APIs, backed by a metric shitton of inline and out-of-line SH4 assembly, SH4ZAM is meant to provide game developers with an ergonomic, high-level way to harness the full potential of the SH4's floating-point and memory performance, without requiring anyone to sell their souls or print the SH4 hardware docs out on their bedsheets (as I've already done that part for you). 🤣 What began as a series of math optimizations I was making to Rockstar's engine and Renderware's middleware layer in our Grand Theft Auto 3 and Vice City ports to improve FPS, wound up sending me on an epic quest to bring back the gainz and democratize them for the Sega Dreamcast community and homebrew scene at-large. Since then, SH4ZAM has grown exponentially more capable and has been leveraged to successfully bring the gainz to high-profile ports such as jnmartin's Mario Kart 64 and Star Fox 64 as well as many homebrew games and engines by several collaborating developers who were willing to work with me during SH4ZAM's early stages, providing invaluable feedback and previewing the gainz. I'd like to personally thank everyone who contributed to the SH4ZAM GitHub repository, for everything ranging from epic inline SH4 assembly routines to contributing bug reports, examples, and documentation improvements. We dun good work, everyone! 🔥 You can add SH4ZAM to your existing KallistiOS development environment by simply pulling down the master branch from the kos-ports GitHub repository and running the "make install" command within the "sh4zam" port directory. Further details and documentation can be found on the README for the GitHub repo: github.com/gyrovorbis/sh4zam
22
63
4
492
23,406
📣 #DreamSDK R4 is out! This is a modern, ready-to-use package for Sega #Dreamcast development on Windows. If you’re on Windows and want to start coding for your Dreamcast right away, this is for you. 📥 Download: github.com/dreamsdk/dreamsdk…
8
167
10
595
64,592
SHE'S DONE!!! Mario Kart 64 for the Sega Dreamcast is officially ready for public consumption, so here's a full 6+ minute video of gameplay, captured straight from my actual DC! Jnmartin has done it again! The footage you are watching shows every GP course in the game, including multiplayer footage as well. Yes, you can also save and load ghost files from the VMU! As this is our first release, I'm sure you will run into a minor bug here or there, especially given neither one of us could stress test 4 player very effectively using our hands and feet to drive... So be sure to open tickets for any minor issues you find... assuming you're playing on REAL HARDWARE. I can already tell you right now, emulators are having issues with this one! This was my first time getting to directly contribute to a jnmartin N64 port, and it was an absolute honor for me to be involved with his incredible work. There is nobody else who I have ever met with the cross-architecture knowledge of BOTH the Sega Dreamcast and Nintendo 64, combined with the raw grit, tenacity, and attention to detail required to pull this off and do the source material justice. Not only did I get to take my SH4ZAM library for DC math gainz out for a ride with MK64, but I also brought some new routines back from MK64 so everyone can use them! Finally, I got to be involved in laying the foundation for future jnmartin N64 ports, now that we have an optimized, accelerated audio and video back-end for various libultra functionality, including Fast3D. You can already see how much this has benefited future work based on how quickly jnmartin was able to get StarFox 64 up and running on DC... The man doesn't sleep, no. lol. I just wanted to thank everyone so much for the continued enthusiasm and support behind these ports. The support for our work and the Sega Dreamcast means the world to us! Thanks for helping to keep the dream alive! For instructions on how to build yourself a burnable CDI image or a DreamShell ISO for either your DC's SD card or IDE drive, please follow the README on the repository: github.com/jnmartin84/mk64-d…. NOTE: Windows users will need DreamSDK to build. My last-minute build fixes for DreamSDK are in this WIP branch here, which will probably be merged into the main branch tomorrow when jnmartin wakes up: github.com/jnmartin84/mk64-d…
66
345
28
1,658
63,719
Guys, I’m too exhausted AF from doing last-minute troubleshooting to post a good final trailer for the alpha version of our Mario Kart 64 port to Sega Dreamcast just yet… YES, the repository went live last night, and you can build it yourself (unless you’re a Windows DreamSDK user, then the build is screwed up, YES I KNOW SORRY SORRY GOTTA SLEEP FIRST BEFORE I CAN FIX). Anyway this right here is the best physical hardware capture I have seen that does the port justice on a real Dreamcast. What do you guys think!?!
4
12
144
5,751
Troy Davis retweeted
Time for a KallistiOS GameCube update! Besides some loose ends, the KOS kernel is ported and all the foundational stuff like threads, context switching, interrupt handling, etc. are working. A bunch of the basic KOS examples/tests now build and run for GameCube. Wrote some framebuffer and joypad code and what else should someone do then if not port Doom over? This is a capture from real hardware. Software rendered and no sound, but it's a start. We are bug squashing so we can release KallistiOS 2.2.0 stable (for Dreamcast) in the upcoming weeks and once that is complete, the process of merging the GameCube arch into KallistiOS will begin! Once that occurs, we hope to have some faster driver development to bring the GameCube arch up to speed with Dreamcast. Ultimate goal is single codebases targeting Dreamcast and GameCube together ❤️
5
14
4
92
13,910
Today we're taking a very early look at the progress being made on the Sega Dreamcast's very own port of Runescape Classic! If you weren't already aware of the work that developer, Lobotomy, has been doing, here's some footage captured from an emulator of some online play! It's still extremely early, and as with every ambitious port in the beginning, it still requires the 32MB RAM upgrade, has no audio, and runs like ass... So basically standard procedure! He's currently using our SDL2 back-end with software rendering, which explains the low framerate. Eventually it will move to HW accelerated graphics using the DC's PowerVR GPU, so the current low FPS is to be expected until that time. What IS a big deal is that even in this early stage, it's already able to connect to the online servers for cross-play and is fully playable with the mouse and keyboard! Stay tuned for future updates as Lobotomy continues his work!
14
31
3
247
11,898
Loved the new video from PCN covering the work we've been doing in the community porting titles from various consoles to the Sega Dreamcast. I will get the pleasure of joining him on a livestream for an interview on the 21st!
The #SEGA #Dreamcast is still alive and kicking. It's getting so many games ported to it, and in most cases it ends up being the definitive version of the game. Video in reply
5
26
205
8,280
Been writing SH4 assembly code for the Sega Dreamcast all day and night, hoping to bring big performance gainz to everyone in the community by providing a replacement memcpy() routine that doesn't suck for our GCC toolchains. As it turns out, the Newlib-provided memcpy() we have backing the C standard library in our SH GCC toolchains is slow AF. This impacts not only our Grand Theft Auto 3 and Vice City ports, but also Doom64, Mario Kart 64, WipeOut, and virtually every homebrew game or port that uses KallistiOS! Just have a look at the benchmark results on the left to see just how shittily it performs. The benchmarker invokes a series of memcpy() implementations over an increasingly large buffer window with compile-time configurable alignments. Each iteration initializes the source buffer with a series of randomly generated numbers and clears the destination buffer before clearing both the data and icaches for each run. During the run, the performance counters on the SH4 CPU are used to record cycle-accurate timing for each memcpy() invocation, which is then validated after the run for correctness. There are also large buffers located before and after the destination buffer, which are scanned for any stray/out-of-boundary writes after each iteration. ANYWAY, what you're seeing in the benchmark output is the performance of my custom 1, 2, 4, 8, and 32-byte aligned memcpy() variants, which are highly optimized for specific use-cases, as well as the result of "memcpy_gainz()" which is the generalized form which attempts to call into the fastest of these specialized forms. Meanwhile, "memcpy_fast()" is a routine we found on the internet many years ago from STMicroelectronics which has impressive speeds, but has an LGPL license, which prevents us from statically linking to it in closed-source commercial games. Finally, "memcpy()" is the C standard library routine that ships with our toolchains... and as you can see... It runs like absolute, total, and complete shit. Somehow, at a pathologically best-case alignment of 32-bytes with 1024-byte+ copy requests, the damn thing manages to be slower than "memcpy1()" which is a simple for loop in vanilla C that could've been written by a total newbie that just copies the source buffer to the destination buffer one byte at a time... So basically all of the bazillion things that are using memcpy() in our software in the Dreamcast community, including everything ranging from copying strings or vertices to transferring packets to and from the layers of our network stack, is all taking a massive performance hit due to us having a shitty memcpy() implementation. After I discovered this, I embarked on a quest to take my specialized memcpyN() routines and see if I could use them as the basis for a generalized memcpy() routine to leverage. This is how "memcpy_gainz()" was born. Unfortunately I was on my own for this quest, as every single resource that I found for writing optimal memcpy() routines was targeted at platforms which support unaligned memory accesses. Such platforms require a fundamentally different approach from the one taken for SuperH and other RISC processors without such support. Rather than simply falling back to unaligned memory accesses, my routine attempts to align the destination buffer to 32-byte cache line boundaries where it can call into one of the fast specialized routines depending on the relative alignment of the source buffer. Then it simply does byte-by-byte unaligned copying for any bytes before or after the cache line boundaries. At this point in time, I'm happy to say that for all alignment types I am beating even our fast_memcpy() implementation for transaction sizes larger than 32 bytes and smaller than 8KB. There's still plenty of work to do for both tiny and massive sizes, but I'm stoked to see what people do with the extra cycles once this is done!
34
71
3
688
53,962
Check out the epic work Troy has been doing with real-time FMV decoding and video playback on the Sega Dreamcast! For video, he's first decoding the source stream into a series of PNG images using ffmpeg. These images are then fed into our Dreamcast PVR texture converter tool (pvrtex), where they are converted into RGB565 format with VQ texture compression applied, before finally being compressed again with the zstd compression algorithm. For audio, ffmpeg is used to convert the input stream into a single, big-ass WAV file, after which it is converted into 4-bit mono ADPCM format using our Dreamcast audio converter tool (dcaconv) we developed for our GTA3 and VC ports. Once video and audio have been encoded, they are combined into a single ".dcmv" (Dreamcast movie) file, which can be loaded by the video player, allowing for playback within a game or app. One of the most interesting issues Troy encountered was trouble with keeping the video and audio streams synchronized when seeking around--the exact same problem that plagued our GTA3 port for quite some time. Due to inherent imprecision in the way time we typically maintain ticks on the SH4's CPU, the time that elapses on the AICA SPU will slowly drift away from it. To work around this, we've had to use the AICA's obscure timer registers directly to maintain consistent frame timing for both CPU-side video and SPU-side audio streams.
Dirk the Daring rides again — now with real-time FMV on the Dreamcast! Source video: 640×480 @ 23.97fps, scaled to 320×240, centered into a 512×256 frame. RGB565 textures VQ-compressed via pvrtex Audio: 32kHz mono ADPCM via dcaconv Zstandard compression: level 22, windowLog 15 Real-time decompression + texture upload via Dreamcast’s SH-4 & PVR DMA 🌀 All decoded live at ~23.97fps on original hardware, synced perfectly to audio 🗃️ .dcmv container holds video + audio + frame index 🔄 CDI bootable image generated gpf@GPF:~/code/dreamcast/dreamcast-fmv/playdcmv$ ls -la movie.dcmv -rw-r--r-- 1 gpf gpf 649619843 Jun 10 11:58 movie.dcmv gpf@GPF:~/code/dreamcast/dreamcast-fmv/playdcmv$ ls -la dcmv.cdi -rw-r--r-- 1 gpf gpf 743007817 Jun 11 12:28 dcmv.cdi 🔗 Code: github.com/GPF/ 💼 Open to new roles: linkedin.com/in/troyedavis #Dreamcast #Homebrew #GameDev #SDL3 #OpenSource #RetroDev #OpenToWork #FMV #JobReferral
5
24
164
6,826
One of the most ambitious engines in the Sega Dreamcast community is now leveraging the DC's 100Mbit broadband adapter for improved asset streaming, load times, and persistent storage, by communicating with a remote file server hosted on the local network!
Dreamcast voxel engine update: Worlds now save and load to disk!
12
76
2
660
16,856
“Vaping causes lung damage”? Not according to CoEHAR. Their new review of 12 studies found no significant changes in lung function—even among daily users. Time to cut through the smoke. #harmreduction clearingtheair.eu/en/post/va…
4
56
7
121
3,230
Replying to @NewYorkStateAG
What are you talking about? Not reality. Either @CDCgov is correct and youth vaping is at the lowest level in a decade, or you are right and there is a nationwide youth vaping crisis. nbcnews.com/health/health-ne…
1
8
1
38
823
I had the honor of getting to fly to Brussels to speak in the GCC Devroom at FOSDEM 2025 on behalf of the Sega Dreamcast homebrew community, talking about all of the amazing things we've done with the platform over this last year, powered by the latest and greatest version of the GCC toolchain! If you're interested in taking a casual stroll through C++20 timezone databases, C++26 concepts/constraints, low-level compiler optimizations, SH4 Assembly, OpenGL graphics with Ada, and much more, featuring Super Mario 64, Sonic Mania, Doom 64, and Grand Theft Auto 3, all running on the Sega Dreamcast... my talk is now live! fosdem.org/2025/schedule/eve…
13
38
2
250
8,916
Is there anything Sega Dreamcast can't do? 🔊 Dreamcast Playa Revival 2025 🔊
11
70
4
495
13,548
A brand new update of all the work @memorix101 @falco_girgis @GPFTroy and Bruce are putting into the port of Carmaggedon for the #Dreamcast youtu.be/cSm-3HnOr68 - Take a peek! #homebrew #dreamcasthomebrew #gamedevs @TheDCJunkyard @timeextension64
10
31
1,060
Is 2025 the Year of new Ports to Dreamcast? In the first #DreamPod podcast of the year, we're rounding up the most exciting homebrew ports, some crazy unreleased ports of Shenmue, as well as the latest releases from the indie and fan translation scenes. 👉buzzsprout.com/42610/episode…
1
20
4
51
11,204
DCJY brings us the full rundown of every single software submission for this year's DreamDisc '24 Sega Dreamcast game jam, including an impressive variety of games, tech demos, VMU apps, and even a custom port of the Java Virtual Machine running on the little white box of dreams... If you haven't already, get yourself some sweet new DC software and help support our developers by checking out the submissions and letting them know what you thought! itch.io/jam/dream-disc-24/en…
The Dream Disc '24 Dreamcast homebrew game jam brought us an excellent variety of new Dreamcast software - from games to VMU apps and more. We have played them all and have summarised each entry on the blog! thedreamcastjunkyard.co.uk/2…
1
14
82
5,674