Connect with us

Tech

Cracking a long-standing weakness in a classic algorithm for programming reconfigurable chips

Published

on

Cracking a long-standing weakness in a classic algorithm for programming reconfigurable chips


Credit: Pixabay/CC0 Public Domain

Researchers from EPFL, AMD, and the University of Novi Sad have uncovered a long-standing inefficiency in the algorithm that programs millions of reconfigurable chips used worldwide, a discovery that could reshape how future generations of these are designed and programmed.

Many industries, including telecoms, automotive, aerospace and rely on a special breed of chip called the Field-Programmable Gate Array (FPGA). Unlike traditional chips, FPGAs can be reconfigured almost endlessly, making them invaluable in fast-moving fields where designing a custom chip would take years and cost a fortune. But this flexibility comes with a catch: FPGA efficiency depends heavily on the software used to program them.

Since the late 1990s, an algorithm known as PathFinder has been the backbone of FPGA routing. Its job: connecting thousands of tiny circuit components without creating overlaps.

For decades, it worked so well that it became the standard. However, as circuits grew larger, engineers began encountering frustrating slowdowns and occasional outright failures. Designs that should have worked were often labeled “unroutable.”

Now, with colleagues from the University of Novi Sad and the technology company AMD, researchers from the Parallel Systems Architecture Laboratory (PARSA) in the School of Computer and Communication Sciences have come one step closer to untangling the inner workings of this classic algorithm.

In their paper, which received the Best Paper Award at the 33rd IEEE International Symposium on Field-Programmable Custom Computing Machines, they revealed why these failures happen and how PathFinder’s limits can be overcome.

Cracks in the algorithm

“In fact, it’s not surprising that PathFinder sometimes fails,” explained Shashwat Shrivastava, Ph.D. student with PARSA and first author of the paper.

“Very early on, researchers showed that the problem behind FPGA routing is extremely hard. Later, the creators of the original algorithm, together with a few collaborators, found cases where PathFinder would never succeed—but they noted such cases wouldn’t appear in practice.”

For decades, it seemed they were correct—PathFinder worked surprisingly well.

“PathFinder worked so well, in fact, that when it failed, people rarely questioned the algorithm. Instead of venturing inside to see what was going on, they tweaked its parameters, modified circuits, or switched to larger FPGAs,” added Stefan Nikolić, an EPFL alumnus and now a professor at the University of Novi Sad.

“Part of the reason for this is that it is rather difficult to understand what PathFinder is actually doing on examples of practical importance. Modern circuits are so large that their signals form veritable on-chip jungles.”

Enter the forest

“So, we really needed to look at the individual trees in that jungle,” continued Shrivastava, “and I really mean trees. Each signal—a connection that carries information between circuit components—must reach multiple destinations without overlapping other signals. FPGA routing is essentially about building one tree for each signal on the chip.”

While working on another project that relied on PathFinder, the team kept seeing results that defied intuition. At first, they blamed external factors, not the itself. Eventually, they realized they needed controlled examples: small, tricky cases where a solution definitely existed, and in which PathFinder should succeed.

“We needed real, practical examples, and lots of them, to understand what was really going on,” Shrivastava explains. “So, we built a framework to automatically extract small, hard problems from real circuits. Watching how PathFinder struggled with these helped us uncover issues that had remained hidden for a very long time.”

Power in partnership

“This breakthrough would have been much harder without industry support,” said Mirjana Stojilović, Shrivastava’s Ph.D. advisor. “From the start, we collaborated with Chirag Ravishankar and Dinesh Gaitonde from AMD. They helped us model FPGAs as close as possible to commercial devices, ensuring our findings had real-world impact.”

Once the framework was ready, things moved quickly. The team found that PathFinder often built routing trees larger than necessary, increasing the risk of overlaps. The problem came from the order in which it created and added new branches to the trees.

“In retrospect, this is intuitive, but somehow it went largely unnoticed for many years,” Shrivastava said. “Our first solution was simple: try different orders and pick the one that results in the smallest tree. Experimentally, it worked surprisingly well.”

The team is now exploring more scalable solutions. “I am especially proud that Summer@EPFL interns have been contributing significantly. One of them, Sun Tanaka, is also a co-author of the paper,” added Stojilović.

“Our discovery could reshape how millions of FPGAs are programmed and influence the design of future generations of these reconfigurable chips.”

More information:
Shashwat Shrivastava et al, Guaranteed Yet Hard to Find: Uncovering FPGA Routing Convergence Paradox, 2025 IEEE 33rd Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM) (2025). DOI: 10.1109/fccm62733.2025.00060

Citation:
Cracking a long-standing weakness in a classic algorithm for programming reconfigurable chips (2025, October 3)
retrieved 3 October 2025
from https://techxplore.com/news/2025-10-weakness-classic-algorithm-reconfigurable-chips.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.





Source link

Tech

Microsoft Has a Plan to Keep Its Data Centers From Raising Your Electric Bill

Published

on

Microsoft Has a Plan to Keep Its Data Centers From Raising Your Electric Bill


Microsoft said on Tuesday that it would be taking a series of steps towards becoming a “good neighbor” in communities where it is building data centers—including promising to ask public utilities to set higher electricity rates for data centers.

Speaking onstage at an event in Great Falls, Virginia, Microsoft vice chair and president Brad Smith directly referenced a growing national pushback to data centers, describing it as creating “a moment in time when we need to listen, and we need to address these concerns head-on.”

“When I visit communities around the country, people have questions—pointed questions. They even have concerns,” Smith said, as a slide showed headlines from various news outlets about opposition to data centers. “They are the type of questions that we need to heed… We are at a moment of time when people have a lot on their mind. They worry about the price of electricity. They wonder what this big data center will mean to their water supply. They look at this technology and ask, what will it mean for the jobs of the future? What will it mean for the adults of today? What will it mean for their children?”

The announcement follows a post from President Donald Trump on Truth Social Monday in which he pledged that his administration would work with “major American Technology Companies,” including Microsoft, to make sure that data centers don’t inflate customer utility bills.

“We are the ‘HOTTEST’ Country in the World, and Number One in AI,” Trump wrote in the post, in which he also accused Democrats of being responsible for the rise in utility bills. “Data Centers are key to that boom, and keeping Americans FREE and SECURE but, the big Technology Companies who build them must ‘pay their own way.’”

Average electricity bills have risen faster than inflation in recent years in many parts of the country. These price hikes are due to a variety of factors, including the costs of repairing and maintaining the country’s aging electric grid. But higher demand for electricity—including from data centers, which can also be expensive to connect to the grid—plays a role. As technology companies and utilities are predicting a massive new need for energy from the nationwide data center buildout, the Energy Information Administration projects that electric bills will keep increasing through 2026.

Concerns around data centers and electricity bills played a key role in several local and state midterm elections last year, while research released last fall shows that local opposition to data centers skyrocketed in the second quarter of 2025, leading to billions of dollars in projects stalled or cancelled. The political divide against data centers appears to be bipartisan. In recent months, influential former Trump strategist Steve Bannon has begun speaking against the energy and water costs of data centers on his War Room podcast, part of a larger pushback from some MAGA figureheads against the AI buildout in the US.

The Trump administration, by contrast, has made expediting the data center buildout in the US a key priority. It has removed a variety of environmental protections for data centers, including water protections, expedited the review of chemicals involved in their use, and encouraged their development on federal land. The Department of Energy has also instructed the Federal Energy Regulatory Commission, which oversees interstate transmission, to work on a suite of issues around data centers and the grid.

Microsoft, which has around 100 data centers planned or under construction across the country, has met with some local pushback to some of its projects. In October, the company canceled plans for a data center in Wisconsin due to local opposition; the group leading the charge against that project warned of a potential “5-15% rate hike to subsidize cheap power.” The company revealed last week that it was also behind a proposed project in Michigan, which was put on hold in December following concerns from community members. Hundreds of residents attended a planning commission meeting for the project Monday night, with many telling local media they were there to express opposition.



Source link

Continue Reading

Tech

Board Is Tabletop Tablet Gaming—but With Physical Pieces

Published

on

Board Is Tabletop Tablet Gaming—but With Physical Pieces


My kids also enjoyed Save the Bloogs, an obviously Lemmings-inspired platformer, where you use the pieces as ladders, bridges, and canons to guide the cute wee Bloogs away from impending death. Board Arcade is the only one that reuses pieces, with spaceships and robots adding physical controls to versions of classics like Snake and Asteroids.

Most of the games are marked as age 6 and up, and I’d expect young kids to love this. The dull digital pet Mushka is clearly aimed at wee ones. Sadly, a couple of the more interesting titles, international mystery Spycraft and strategy face-off Thrasos, are still marked as “coming soon.” The Tetris and Chess-inspired Strata is a bit more challenging and makes for a fun two- to six-player battle for territory. I also quite liked the sushi-themed Omakase, a head-to-head battle to build the best bento boxes.

The quality and depth of the launch titles vary, but they feel more like examples of what you could do with the tech than must-have games. It’s telling that the best of the bunch are clones. It will be interesting to see what third-party developers might come up with. While I applaud the lack of a subscription, it’s not yet clear how much the 10 new incoming titles will cost, when they will land, or how many more might follow. You’ll be able to download the titles digitally, and the physical component packs will be shipped.

Mixed Reception

I really like Board, but my teen kids (13 and 16) weren’t impressed. They both found Board’s games boring. Other family and friends had similarly mixed reactions. We play a lot of video games and board games, but I was the only one pushing Board over the holidays. It lacks that killer addictive hook that keeps you coming back for more. I suspect it would go down a lot better with younger kids (6 to 10).

Board is big, and I struggled to find a spot to store it. While it has survived mostly unscathed so far, I can’t help wondering about long-term durability, as the frantic nature of some games could lead some, especially younger kids, to be a bit rough with the screen. It has already sustained a faint scratch from a wayward cat claw, as one of my cats has the unfortunate habit of sitting on board games when we gather to play.



Source link

Continue Reading

Tech

For Under $100, Atonemo’s Streamplayer Makes Old Speakers New Again

Published

on

For Under 0, Atonemo’s Streamplayer Makes Old Speakers New Again


Photograph: Chris Haslam

Plugged into my father’s 20-year-old Arcam amplifier, powering a pair of equally vintage Mordaunt Short floorstanding speakers, the Streamplayer works smoothly. It did require a 3.5mm to RCA cable (more in this below) but dad’s box of old cables is a tech treasure trove. On first listen I was impressed by the ease of it all, and switching between Qobuz and Spotify Lossless certainly didn’t offend. But when playing the same song side-by-side using a mid-range Cambridge Audio CD player, the difference was obvious.

Playing the album “Music from Big Pink” by The Band, and the power and scale from the CD version easily eclipsed that of the Streamplayer. Rick Danko’s bass lines soared and the layering of the instruments was wonderfully pronounced via CD, while they were noticeably subdued when streaming. Same speakers, same amplifier, very different DACs.

It’s by no means a disaster, and the streaming convenience—in lieu of a large CD collection—is difficult to argue with. Even with quality components, the performance is only ever going to be as good as the audio quality played.

Cables and Connectivity

Atonemo makes a big deal over being able to power “all” legacy speakers. Included in the box is a single 3.5mm to 3.5mm audio jack, giving the ability to plug into active speakers, old radios, cassette players, boomboxes, and anything with a 3.5mm aux-in port.

But to power analogue speakers, you will need a separate amplifier and alternative cables, whether that’s a 3.5mm to RCA or SPDIF RCA to 3.5mm jack cable. Atonemo told WIRED they toyed with the idea of including multiple cables in the box, but opted for the 3.5mm audio jack because most hi-fi people would have one already. They also plan to sell a range of cables to suit “all” legacy speakers. I think including a couple of cable options would have been a good idea, especially at launch, just to avoid any friction points with first-time installation, as not everyone has a box of old cables as comprehensive as my Dad.

Competition

Image may contain Electronics and Speaker

Photograph: Chris Haslam

Atonemo isn’t the only option when it comes to breathing new life into older hi-fi components. The WiiM Mini Music Streamer ($89) is cheaper, offers AirPlay/Chromecast/Spotify Connect streaming capabilities and a XX DAC. Similarly the FiiO SR11 Desktop Streaming Music Receiver ($110) also copes with Apple Music, Spotify, Tidal, FiiO Music and PCM 768 kHz/32 Bit and DSD256 (DOP) quality. And for a further boost in audio quality plus one of the finest hi-res streaming multiroom platforms, the $379 Bluesound Node nano streamer is hard to ignore. There are also plenty of basic Bluetooth dongles available too if audio quality, or the ability to enjoy multi-room audio, isn’t important to you.

But what Atonemo has done well is remove any technical barriers to use. The app is incredibly simple—in a good way—with no bloat, or pretence it is doing anything other than facilitating streaming between your old speakers and a modern streaming platform.

Yes, it could (and probably should) squeeze in a better DAC, and a more comprehensive collection of cables, but these are deliberate omissions, not glaring mistakes. Anyone looking to tweak and tune, and generally nerd-out about bit rates, are already well served elsewhere.



Source link

Continue Reading

Trending