Intel Celeron N5095 vs Intel Core m3-8100Y

Intel Celeron N5095

The Celeron N5095 is an inexpensive quad-core SoC of the Jasper Lake product family designed for use in affordable SFF desktops and laptops. It features four Tremont CPU cores running at 2 GHz that Boost to up to 2.9 GHz with no thread-doubling Hyper-Threading technology in sight. A pretty basic iGPU is present as well.

The only difference between the N5095 and the N5095A is that the latter comes with support for more proprietary Intel technologies such as the Smart Sound DSP, Wake on Voice and HD Audio.

Architecture and Features

Tremont brings many improvements over Goldmont Plus, the architecture that we know from the N5030 and myriads of other N-class CPUs. An up to 30% boost in single-thread performance is to be expected thanks to smarter prefetchers, branch prediction improvements and other refinements, according to Intel. These new chips are physically larger than their immediate predecessors as a result. Either way, this is still a "small" core rather than a "big" one according to ChipsAndCheese.

The Celeron has 1.5 MB of L2 and 4 MB of L3 cache and is compatible with DDR4-2933 and LPDDR4x-2933 memory or slower. Support for Intel CNVi Wi-Fi 6 modules is baked into the chip, as are 8 PCIe 3.0 lanes for NVMe SSD speeds up to 3.9 GB/s. USB 4 or Thunderbolt aren't supported however.

Please also note that the Celeron gets soldered to the motherboard (BGA1338 socket interface) for good and is thus not user-replaceable.

Performance

The average N5095 in our database is about as fast as the Core i3-10110U, Core i3-1005G1, Celeron N5105 and also the Ryzen 3 3200U in multi-threaded workloads. Which is just enough for the most basic of tasks in late 2024.

Performance will get a significant hit if the power target is set to 10 W or 6 W instead of the Intel-recommended 15 W value.

Graphics

The DirectX 12.1-capable 16 EU UHD Graphics runs at up to 750 MHz and is in many respects similar to what Ice Lake CPUs come equipped with. This graphics adapter is capable of driving up to 3 SUHD displays simultaneously; HEVC, AVC, VP9, MPEG-2 and other popular video codecs can all be hardware-decoded. AV1 and VVC can't.

As far as gaming is concerned, it is reasonable to expect playable framerates in really old games (like Dota 2 Reborn) provided one sticks to lower resolutions such as HD 720p.

Power consumption

While most N-class chips have a 6 W long-term power target, the Celeron N5095 has a 15 W TDP to mimic much faster U-class Core processors. This isn't a great CPU for passively cooled designs.

The N5095 is built with the same 10 nm Intel process as Ice Lake-U processors for pretty unimpressive power efficiency, as of late 2024.

Intel Core m3-8100Y

The Intel Core m3-8100Y is a very efficient dual-core SoC for tablets and passively cooled notebooks based on the Amber Lake generation and will be announced around August 2018. The CPU consists of two processor cores clocked at 1.1 - 3.4 GHz. Thanks to Hyper Threading, the processor can execute up to four threads simultaneously. The chips also includes the Intel HD Graphics 615 GPU, a dual-channel memory controller (DDR3L/LPDDR3) as well as VP9 and H.265 video de- and encoder.  Compared to the Kaby Lake-Y predecessors (e.g. Core m3-7Y32), Amber Lake uses the same architecture produced in the improved 14nm+ FinFET process (no 14nm++) but offers higher clock speeds at an increased TDP rating.

Architecture

Intel basically used the familiar micro architecture from the Skylake / Kaby Lake generation, so the per-MHz performance is identical.

Performance

The performance of the Y series is highly depending on the cooling solution of the laptop / tablet and the TDP settings for sustained load. Therefore, the differences between a Core m3 and a i7 may be rather slim, if the thermals are very limited. The high Turbo clock speeds in Amber Lake should however lead to a very good single thread performance for short bursts.

Graphics

The integrated Intel HD Graphics 615 GPU has 24 Execution Units (EUs) like the old HD Graphics 515 and runs with clocks between 300 and 950 MHz in combination with this processor. The performance heavily depends on the TDP limit as well as the memory configuration; with fast LPDDR3-1866 RAM in dual-channel mode, the GPU should sometimes be able to compete with the HD Graphics 520, but can also be much slower in other scenarios. Modern games from 2016 will, if at all, only run smoothly in the lowest settings.

Contrary to Skylake, Kaby Lake and Amber Lake now also supports hardware decoding for H.265/HEVC Main10 with a 10-bit color depth as well as Google's VP9 codec.

Power Consumption

The chip is manufactured in the old 14 nm+ process with FinFET transistors. The typical TDP for the Amber Lake Y-series is specified at 5 Watts (previous generations were 4.5 Watt), and can be adjusted in both directions depending on the usage scenario.