Tim did a great write up of the new hardware found in the Raspberry Pi 3 Model B+ which you should check out below if you missed. Technical specifications are only the first step as we still need to see how the new 1.4GHz Cortex A53's perform in benchmarks and Phoronix have published just that. They compared the Pi 3 to a variety of chips including the previous model, ASUS' Tinkerboard, the two Jetson boards, a few Celerons and even a Core i3. Overall the chip showed an advantage over the previous model; not earth shattering but as the price remains at $35 for the Pi 3 that is still a good deal.
"I've been spending the past few days putting the Raspberry Pi 3 Model B+ through its paces the past few days with an array of benchmarks while comparing the performance to other ARM SBCs as well as a few lower-end Intel x86 systems too. Here is all you need to know about the Raspberry Pi 3 B+ performance."
Here is some more Tech News from around the web:
- 2 + 2 = 4, er, 4.1, no, 4.3… Nvidia's Titan V GPUs spit out 'wrong answers' in scientific simulations @ The Register
- Best Buy Stops Selling Huawei Smartphones @ Slashdot
- Apple to enter trial production of new iPhone series in 2Q18, say sources @ DigiTimes
- ICO still waiting for 'urgent' warrant to raid Cambridge Analytica's London HQ @ The Inquirer
- Mozilla Pulls Advertising from Facebook @ Slashdot
- Facebook's Zuck comes out of hiding, admits company 'made mistakes' @ The Inquirer
- Seagate's HAMR to drop in 2020: Multi-actuator disk drives on the way @ The Register
- Slack's GDPR changes means admins can now snoop on private chats @ The Inquirer
- Tomb Raider Remasters Have Been Cancelled @ [H]ard|OCP
- HITMAN Spring Pack Is FREE For A Limited Time! @ Tech ARP
Please keep in mind that the
Please keep in mind that the benchmarks done at Phoronix are horribly flawed and need to be taken with a few pounds of salt. For example, any of the storage benchmarks are effectively random numbers as he uses random uSD cards that he finds lying around to run his boards. uSD cards can easily vary by over a factor of 100 in performance depending on the workload. Some of the more PC like boards he tests have full up SSDs in them.
The OS versions he chooses are equally flawed and seemingly random. There are several of his results which are clearly incorrect–where the Rpi3+ performs more slowly than the Rpi2.
People point out the issues with his testing methodology time and again, but he doesn’t listen–choosing to repeat the same mistakes over and over again.
It’s very saddening because he means well and his testing suite could be much more useful if he listened to the feedback he’s getting, but it falls on deaf ears.
I appreciate the work he does, but his results are so tainted that he ends up doing more harm than good.