NZCity NewsLinks -

All Links

13 Mar 2025

Search

Casino advertising approval prompts problem gambling fears
"Allowing advertising and legitimising operators through licensing could have the unintended effect of encouraging more gambling," officials warn.

Master Electricians' warning on the Right to Repair Bill
The bill passed its first reading in Parliament last month, but one trade feels the amendment is a live wire.

Cheating scandal shocks ski jumping, topples Olympic champions and shakes Norway's lofty reputation
Cheating by Norway team officials manipulating ski suits has shaken a national reputation for fair play and high-minded principles.

In wake of scandal, Google clamps down on Chrome shopping extensions
Late last year, there was a big scandal involving the Honey browser extension, which wasn’t helping users find the best deals and coupons as it so claimed to do. Instead, it was swapping out affiliate links and stealing potential commissions from influencers and creators. Following that revelation, Google is now tightening the rules for shopping extensions in the Chrome Web Store. In short, affiliate links, coupon codes, and cookies can now only be included in the extension if they provide a direct and clear benefit to users when they shop. For example, extensions can’t insert links unless they lead to real discounts or cash back on purchases. Extensions also can’t update your shopping cookies or inject affiliate links on web pages without explicitly notifying you. While the Honey scandal mainly affected influencers and creators — not Joe Schmoe who’s just browsing the web — this is generally good news as you can be more sure that extensions are doing what they claim. Further reading: I hope Google does lose ownership of Chrome

Addlink S93/A93 SSD review: Good value if you skip the heatsink
At a glanceExpert's Rating Pros PCIe 4.0 host memory bus performance Heatsink-less S93 is a super bargain Five-year warranty Cons One of the slower drives in its class Our Verdict The Addlink A93/S93 is a solid-performing PCIe 4.0, DRAM-less NVMe SSD that can save you a few bucks. But the A93 with its heatsink faces heavy competition from many similarly priced and often faster SSDs. Price When Reviewed This value will show the geolocated pricing text for product undefined Best Pricing Today PCIe 4.0 host memory buffer (HMB) SSDs like the Addlink S93 / A93 (heatsink) are all the rage these days. Without the cost of DRAM primary cache added to the mix, the modules are cheaper to produce and can still match DRAM-designs in most performance areas. That said, the 1TB version of the S93/A93 I tested ranks low among a list of very fast SSDs — primarily due to a very slow time in our 450GB write compared to the 2TB competition. Read on to learn more, then see our roundup of the best PCIe 4.0 SSDs to for comparison. What are the Addlink S93’s features? The S93 is a 2280, PCIe 4.0 x4 (four lane) NVMe 2.0 SSD sporting a Phison P27T controller and 162-layer TLC NAND. It’s a host memory buffer design, which means in lieu of DRAM, your device’s memory is used for primary caching duties. When it first showed up, HMB designs were slower than DRAM in most areas. Now they regularly compete well in terms of sequential throughput, though DRAM still rules when it comes to random 4K operations. The gap seems to be slowly closing though. Addlink warranties the S93/A93 for five years and the drives feature a 300TBW per every 1GB of capacity rating. That’s a bit on the parsimonious side — actually, half of what we normally see from TLC drives. The S93 is bargain, the A93, not quite so much. The back of the S93 NVMe SSD. TBW indicates the amount of data that may be written under warranty. Note, that you may be able to write a lot more than that before the drive turns read-only. This is basically a vendor risk calculation. How much is the S93/A93? The S93 we tested is very affordable at $68.44 for the 1TB, $133.44 for the 2TB, and $268.44 for the 4TB. The heatsink on the A93 adds more bucks than I was expecting, making the 1TB version $84.44, the 2TB $149.44, and the 4TB capacity $312.44. The S93 is bargain, the A93, not quite so much. It’s a nice heatsink, but few users need one and the graphene heat spreader/label on the S93 dissipates a fair amount of heat. How fast is the Addlink S93? The 1TB S93 Addlink proved mostly competitive performance-wise, if not up there with the best we’ve seen. What dragged down its overall performance to 24th out of 29 PCIe 4.0/HMB/TLC SSDs that I’ve tested was its miserable 450GB write time. That’s primarily thanks to having only half the NAND of its competitors available for secondary caching. All three PCIe 4.0/DRAM-less/TLC competitors listed in the chart — the Lexar Play 2280, WD Black SN7100, and the Teamgroup T-Force A440 Lite — are 2TB models. As you can see, CrystalDiskMark 8 rated the S93’s sequential transfer abilities highly when queues were in play. But the single-queue performance was more than a bit off the highly ranked WD Black SN7100’s. As you can see, CrystalDiskMark 8 rated the S93’s sequential transfer abilities highly when queues were in play. But the single-queue performance was a bit off the competition. Longer bars are better. Random performance in CrystalDiskMark 8 was better, and actually comparable with a single queue in play. The Lexar’s tragic numbers in these tests were largely a matter of subpar caching, as they turned competitive when we reduced the CrystalDiskMark 8 data set to 16GB. Random performance in CrystalDiskMark 8 was a bit better, and actually comparable with a single queue. Longer bars are better. I have nothing to complain about with the A93/S93’s 48GB transfer times. It’s good in common real-world scenarios. I have nothing to complain about with the S93’s 48GB transfer times. Shorter bars are better. The 450GB is where being only 1TB in capacity hurt the S93/A93 — there’s simply less NAND to treat as SLC secondary cache. However, the slowest write speed was still a SATA-like 600MBps, so this isn’t quite as tragic as you might think in the grand storage scheme. The 450GB is where being only 1TB in capacity hurt the S93 compared to the competition — there’s simply less NAND to treat as SLC secondary cache. Shorter bars are better. Again, in its 2TB flavor, the S93/A93 would likely have ranked a good deal higher. Note that originally I was set to test the heatsink-clad A93 that Addlink also provided. Alas, to use the SSD in my upside-down secondary PCIe 5.0 M.2 slot, the heatsink had to go. Rather startlingly, the two NAND chips came off the PC board along with the heatsink and its thermal coupling material. This should not happen, and was a first for me after performing numerous other similar operations. Make sure you opt for the bare S93 if you don’t need or can’t use a heatsink. Should you buy the Addlink S93/A93? The S93/A93 is hardly a barn burner in the category, but it will do the job, and in its S93 incarnation it’s outstandingly affordable. That said, I recommend the 2TB version and looking for it on sale. How we test Drive tests currently utilize Windows 11, 64-bit running on an X790 (PCIe 4.0/5.0) motherboard/i5-12400 CPU combo with two Kingston Fury 32GB DDR5 4800MHz modules (64GB of memory total). Both 20Gbps USB and Thunderbolt 4 are integrated to the back panel and Intel CPU/GPU graphics are used. The 48GB transfer tests utilize an ImDisk RAM disk taking up 58GB of the 64GB of total memory. The 450GB file is transferred from a 2TB Samsung 990 Pro which also runs the OS. Each test is performed on a newly NTFS-formatted and TRIM’d drive so the results are optimal. Note that in normal use, as a drive fills up, performance may decrease due to less NAND for secondary caching, as well as other factors. This can be less of a factor with the current crop of SSDs with far faster late-generation NAND. Caveat: The performance numbers shown apply only to the drive we were shipped and to the capacity tested. SSD performance can and will vary by capacity due to more or fewer chips to shotgun reads/writes across and the amount of NAND available for secondary caching. Vendors also occasionally swap components. If you ever notice a large discrepancy between the performance you experience and that which we report, by all means, let us know.

Scott Watson appears before parole board
Watson was convicted of murdering Ben Smart and Olivia Hope in the Marlborough Sounds in 1998.

This nightmarish $35K computer is powered by a lab-grown human brain
An Australian company called Cortical Labs has developed a computer powered by lab-grown human brain cells, Gizmodo reports. The computer, known as CL1, is described as the world’s first “code deployable biological computer” and is now available for pre-order — for a price in the $35,000 range. Don’t want to buy your own device? The company also offers “Wetware-as-a-Service” via which you can rent bio-computer processing power via the cloud. CL1 consists of lab-grown neurons grown on a glass-and-metal electrode array. They’re connected to 59 electrodes, creating a stable neural network. The system is encased in a life support unit that keeps the neurons alive by mimicking the body’s organ functions, including heart pumping, kidney-like waste filtration, and gas mixing of oxygen, carbon dioxide, and nitrogen. According to Cortical Labs, the neurons are placed in a nutrient solution and receive their information from the company’s Biological Intelligence Operating System (biOS), which creates a simulated world in which the neurons receive sensory input and produce responses that affect the environment. CL1 is designed as a high-performance closed loop, where neurons interact with software in real time. The system can stay alive for up to six months and is compatible with USB devices. Cortical Labs demonstrated an early version of the technology by teaching the system to play Pong. They claim that biological computers can rival or surpass digital AI systems, especially when it comes to understanding the basic mechanisms of intelligence. According to the company’s Chief Scientific Officer, Brett Kagan, a network of 120 CL1 devices could give researchers insight into how genes and proteins affect learning. The technology can also be used in drug development and disease modeling by simulating neurological processes at the molecular level.