Electronic design automation (EDA), silicon IP and engineering simulation. SNPS / CDNS. Run 8 October 2026. Built from both companies' 10-Ks for FY2023 to FY2025, the latest 10-Qs (Synopsys Q3 FY2026, Cadence Q2 2026), 2025-2026 earnings-call transcripts, 8-Ks and Synopsys's 30 September 2026 Investor Day materials in the Research folder, plus results releases and financial supplements from the companies' IR sites and one TrendForce share estimate (web, labelled). Events swept through 8 October 2026; most recent event checked: Synopsys's $1 billion accelerated share repurchase (8-K, 5 October 2026). Cadence reports Q3 2026 on 26 October 2026. US dollars as reported, no FX conversion. Fiscal years end 31 October (Synopsys) and 31 December (Cadence), a two-month offset. Not a valuation and not a recommendation.
Scale: after buying Ansys in July 2025, Synopsys guides FY2026 revenue of about $9.7B against Cadence's about $6.3B, so this page compares ratios and growth rates, not absolute size.
Three cells carry Cadence's next five years. In Core EDA sold to AI and hyperscaler design teams, agentic add-ons and digital full-flow wins sit on top of an emulation franchise built on Cadence's own processor chips (Q4 2025 call); Core EDA grew 18% in each of the first two quarters of 2026. In silicon IP, a multi-foundry world multiplies the titles each customer needs: Cadence signed a multi-year Intel 14A engagement and deepened Samsung 2nm and Rapidus work (Q2 2026 call), and is winning from a base under half of Synopsys's. In system analysis, 3D-IC packaging tools plus the Hexagon D&E structural suite (closed 23 February 2026, Q2 2026 10-Q) extend the same customers into physics. Synopsys's forward mechanism is real but slower and still a promise: Ansys guided to $2.98B with double-digit growth, an IP rebuild targeted at 17%+ a year, and a FY2027 revenue guide of $11.1B to $11.2B, about 15% (Investor Day, 30 September 2026).
Cadence converts growth through mix and operating leverage: ratable software dominates, IP is 15% of revenue, and the CFO puts organic incremental margin near 60% (Q1 2026 call), so non-GAAP operating margin reached 44.6% in 2025 while absorbing record hardware shipments. Synopsys's gross margin sits about 7 points lower in every year because a quarter of revenue came from people-intensive Design IP, whose segment adjusted margin was 24% in FY2025 against 42% in Design Automation (FY2025 10-K, Note 19), and its stock compensation ran 12.7% of revenue against Cadence's 8.6%. Synopsys guides non-GAAP margin from 37.3% to about 44% in FY2027 and about 50% by FY2030, built on $400M of cost synergies and a 10% workforce cut (Investor Day). The level and the quality of conversion favor Cadence; the slope favors Synopsys, which is why this is narrow.
Cadence's outperformance is concentrated in its most cyclical revenue. Up-front revenue (hardware, IP licences, multi-year system-analysis licences) rose from 17% of revenue in 2024 to 22% in Q2 2026, and China rose to 15% of quarterly revenue while Cadence serves probation under a July 2025 export-control plea, and a BIS rule extending restrictions to 50%-owned affiliates is suspended only until 9 November 2026 (FY2025 10-K, Q2 2026 10-Q). A pause in AI capex or a renewed China restriction hits that slice first, and at a roughly 60% incremental margin the loss falls almost straight to operating profit. Synopsys's exposures are different: $10.0B of gross debt after Ansys against $3.6B of cash (Q3 FY2026 10-Q), and an IP business that showed in 2025 how one foundry customer and two late titles can cut a segment margin from 38% to 24%. But recurring revenue of 82%, China at 10% of revenue and Ansys's industrial customer base make Synopsys's revenue line the harder one to break in an AI-led downturn (Q3 FY2026 supplement and 10-Q).
These five cells cover essentially all revenue of both companies; product rows and region rows are two cuts of the same revenue, because neither company discloses a product x region cross-tab. Every score is argued in the tabs below.
| Cell (product x region) | Synopsys | Cadence | Why (one clause, sourced) |
|---|---|---|---|
| Core EDA, US-led AI/HPC and hyperscaler design teams | 3 | 4 | Cadence Core EDA +18% vs Synopsys EDA +8.5% organic in the latest quarter (calls); TrendForce has Cadence 32% vs Synopsys 30% EDA share for 2025 (web). |
| Silicon IP, AI/HPC and foundry enablement, global | 3 | 4 | Synopsys still sells 2.4x Cadence's IP but fell 8% in FY2025; Cadence grew ~25% and then over 40% in Q2 2026, mostly organic (10-Ks, calls). |
| Simulation and system analysis, industrial US and Europe | 4 | 3 | Ansys is $2.98B and growing double digits; Cadence SD&A is ~$0.85B plus ~$0.2B of D&E, a challenger winning displacements (guide, calls). |
| China, all products | 2 | 3 | Synopsys ex-Ansys fell 22% in FY2025 with share moving to local vendors; Cadence grew 19% and rebounded 64% in H1 2026 off a restricted base (10-K, 10-Q, Q4 FY2025 call). |
| Asia ex-China and rest of world, all products | 4 | 4 | Synopsys Korea +22.5% in FY2025 on HBM and foundry work; Cadence Other Asia +29% and Japan +28% in H1 2026 (10-K, 10-Q). |
Scores are anchored to the exhibits in the three tabs and are not summed; the three lenses get their verdicts in the Three Answers above.
Normalization first. Synopsys reports two segments (Design Automation, Design IP) and four product groups (EDA, Design IP, Ansys, Other); Cadence reports three product categories as percentages only (Core EDA, Semiconductor IP, System Design and Analysis). They map cleanly to three normalized products: Core EDA, Silicon IP, and Simulation and system analysis. Cadence's category includes PCB and package design, which Synopsys largely does not sell, so that row is a looser match. Regions map to United States, Europe (Cadence's EMEA), China, and Asia ex-China plus rest of world (Synopsys Korea and Other; Cadence Other Asia, Japan and Other Americas). Cadence product revenue in dollars is the disclosed percentage times total revenue, which carries about plus or minus $26M of rounding (inferred).
The headline contrast is organic. Synopsys's reported growth is dominated by Ansys, closed 17 July 2025; on the pro forma basis its 10-Q provides, revenue grew 7.4% in the first nine months of FY2026, or roughly 9% after adding back about $150M of divested revenue (inferred). Cadence's pro forma growth, which strips out Hexagon D&E, was 16.7% for the first half of 2026.
| Cell | Synopsys rev (% of total) | Growth | Cadence rev (% of total) | Growth | Margin signal / leader |
|---|---|---|---|---|---|
| Core EDA | $4,370M (62%) | +7.3% FY25; +8.5% Q3 FY26 | ~$3,708M (70%) | +13% FY25; +18% H1 26 | Synopsys Design Automation adj. margin 42% FY25 (segment); Cadence ND by product. Cadence leads on growth. |
| Simulation and system analysis | $757M FY25 (11%, 3.5 months); $2,980M FY26E | double digit FY26E | ~$848M (16%) | +13% FY25; +37% Q2 26 incl. D&E | Synopsys leads by 3 to 4x in size; D&E runs at a 5% to 10% margin in 2026 (Cadence CFO). |
| Silicon IP | $1,752M (25%) | -8.1% FY25; +10.8% Q3 FY26 | ~$742M (14%) | ~+25% FY25; over +40% Q2 26 | Synopsys IP adj. margin 38% FY24, 24% FY25, 26.5% Q3 FY26; Cadence ND. Cadence leads on momentum. |
| United States | $3,100M (44%) | +13.2% FY25 (incl. Ansys) | $2,311M (44%) | +7% FY25; +8% H1 26 | Same weight in both; Cadence's US growth is the slowest of its regions. |
| Asia ex-China and RoW | $2,251M (32%) | +26% FY25 (incl. Ansys) | $1,515M (29%) | +25% FY25 | Even; Synopsys Korea strongest single region (+18% 9M FY26). |
| Europe | $889M (13%) | +45% FY25 (Ansys) | $791M (15%) | +13% FY25; +22% H1 26 | Ansys makes Europe a simulation cell for Synopsys. |
| China | $814M (12%) | -17.7% FY25; ex-Ansys -22% | $680M (13%) | +19% FY25; +64% H1 26 | Cadence leads; its 2026 rise is partly a rebound from the May to July 2025 licence requirement. |
Sources: Synopsys FY2025 10-K (Notes 5 and 19, MD&A), Q3 FY2026 10-Q and supplement; Cadence FY2025 10-K and Q2 2026 10-Q; growth rates for Cadence products from the Q4 2025, Q1 2026 and Q2 2026 calls.
Core EDA. Synopsys EDA grew 7.3% in FY2025, held back by China and by a 53rd week in the prior year, and 8.5% organically in Q3 FY2026 against a 16% comparison; management guides double-digit organic EDA growth for FY2026, with hardware-assisted verification at record levels and 3D-IC Compiler wins such as AMD's MI455X (Q3 FY2026 call). Cadence's Core EDA grew 13% in 2025 and 18% in each quarter of 2026, driven by record Palladium Z3 and Protium X3 hardware, 25 new digital full-flow logos in 2025, a hyperscaler's first full customer-owned-tooling tape-out on Cadence, and early agentic tools such as ChipStack with more than 20 customer engagements (Q4 2025 and Q2 2026 calls). Both are organic.
Silicon IP. Synopsys's IP segment fell 8% in FY2025 after China restrictions, "weaker than expected demand from a major foundry customer", and "roadmap and resource decisions that did not yield their intended results" (FY2025 10-K). It sold its processor IP business to GlobalFoundries on 1 June 2026, returned to growth in Q3 FY2026 (+10.8%), and is moving custom work to a licence-plus-royalty model it calls Factory 2, targeted at about $1B by FY2030 (Investor Day). Cadence's IP grew about 25% in 2025 and over 40% in Q2 2026, which the CEO calls mostly organic, on HBM, UCIe, PCIe and LPDDR6 titles plus a significant Intel agreement (Q2 2026 release and call).
Simulation and system analysis. Both positions are acquired. Synopsys paid about $34.9B for Ansys and guides $2.98B of Ansys revenue for FY2026 with double-digit growth, $400M of revenue synergies by FY2029 and first contributions from joint Multiphysics Fusion products in FY2027 (FY2025 10-K, Investor Day, Q3 FY2026 call). Cadence grew system analysis 13% in 2025 including BETA CAE, then added Hexagon D&E for $2.9B net, about $200M of annual revenue at a 5% to 10% margin in its first year (Q2 2026 10-Q, Q4 2025 and Q1 2026 calls).
China and Asia. Synopsys's China revenue excluding Ansys fell 22% in FY2025, and the CEO described share moving to local vendors "at an accelerated rate" among customers it cannot serve (Q4 FY2025 call). Cadence's China rose 19% in 2025 and 64% in the first half of 2026, partly a rebound from the May to July 2025 licence requirement (Q2 2026 10-Q). In Asia outside China both grow in the mid-20s, with Synopsys Korea up 22.5% in FY2025 on HBM and memory work.
Both sell EDA and IP through direct sales forces on time-based licences of generally two to three years, so neither depends on a distributor for net price (both FY2025 10-Ks). The difference is in simulation: Synopsys also sells Ansys products through independent channel partners and, in valuing Ansys, assumed customer retention of 85% to 105% in the direct channel against 70% to 90% in the indirect one (FY2025 10-K, critical accounting estimates). Neither company splits growth into price and volume. Cadence's CFO said "the overall pricing environment has improved" (Q1 2026 call), while Synopsys spoke of new "pricing guardrails" in IP after its 2025 problems (Q4 FY2025 call). Both plan to charge for AI agents on top of the base tools.
| Input | Synopsys | Cadence |
|---|---|---|
| Emulation hardware | Risk factor cites "dependence on a sole supplier for certain hardware products" | Single or limited suppliers and contract manufacturers; designs its own emulation chips made by TSMC (10-K, Q4 2025 call) |
| Inventory, FY2025 | $365.2M | $303.5M |
| Engineers (the binding input) | ~28,000 staff at FY2025 end, ~25,345 by Q3 FY2026 after a 10% cut | ~13,800 at end-2025; ~15,000 including D&E (CEO, Q1 2026) |
| Foundry certification | Both certified at TSMC, Samsung and Intel nodes | Same; adds Rapidus agentic collaboration |
Supply is not the industry's defining issue. Both carry single-source hardware risk, and Cadence's custom emulation silicon makes its hardware harder to copy but ties it to one foundry. Gross margin held through the 2025 memory-price rise at Cadence (86.0% in 2024, 86.4% in 2025), so pass-through looks adequate.
TrendForce's 2025 forecast puts Cadence at 32%, Synopsys at 30% and Siemens EDA at 13% of EDA revenue, with others at 25% (web, 13 November 2025). Customers' in-house tools are the other named competitor in both 10-Ks, and in China domestic vendors are taking the customers that US export rules close off (Synopsys Q4 FY2025 call). In simulation, Cadence's own 10-K names Ansys, now inside Synopsys, as a key competitor.
Core EDA: design starts thin as chips get costlier, which Cadence's 10-K warns has happened before; Cadence is more exposed because more of its growth is AI-driven hardware. Silicon IP: revenue is recognized per design at delivery and depends on a few foundry roadmaps; Synopsys is more exposed, having already been hit by one. Simulation: industrial and automotive R&D budgets cycle differently from chips; Synopsys is more exposed by size, and its channel customers churn faster. China: export rules; Cadence is more exposed in share of revenue (15% vs 10%) and through its probation terms.
Signoff and implementation at the leading node (High): Fusion Compiler and PrimeTime claimed in use on 100% of critical tape-outs at 2nm and below (Q1 FY2026 call, company claim). The broadest silicon IP catalogue (Medium): $1.75B of IP revenue, 2.4x Cadence's, ported per foundry and per node, which takes years of R&D; 2025 showed it can still lose titles. Ansys multiphysics signoff (High): foundry-certified power and electromagnetic analysis plus the leading industrial simulation suite, which Cadence would need years to match.
Custom and analog design (High): Virtuoso's franchise, now extended with AI-driven migration (Q4 2025 call). Emulation on its own silicon (High): Palladium runs on processors Cadence designs and has made at TSMC, a capability that took years to build and that set records in each of the last three years (Q4 2025 call). Digital full flow (High and rising): 25 new full-flow logos in 2025 and share gains in signoff, which pushed TrendForce's 2025 estimate to 32% vs 30%.
Synopsys: one customer and its subsidiaries supplied 12.4% of revenue in FY2023 and 12.6% in FY2024, and a "major foundry customer" drove the 2025 IP shortfall (FY2023 and FY2025 10-Ks; the filings do not name it, and the Q4 FY2025 call ties it to advanced-packaging IP, consistent with Intel, inferred). Cadence: no customer reached 10% of revenue in 2023, 2024, 2025 or the first half of 2026 (10-Ks, 10-Q). Backlog tells the same story: Cadence's grew from $7.8B to $8.1B in the first half of 2026 with 58% due in 12 months, while Synopsys's fell from $11.4B to $10.9B, partly from the processor IP sale (10-Q, Q3 FY2026 call).
Synopsys: hardware relies on "a sole supplier for certain hardware products" (FY2025 10-K); the binding input, engineers, is being cut by 10% while hiring in priority areas (Q2 FY2026 call). Cadence: single or limited suppliers and contract manufacturers for hardware, with past delivery delays disclosed (FY2025 10-K), offset by owning the emulation chip design. Both pass through input inflation, since gross margins barely moved in 2025.
| Cell | Route control | Pocket price | Continuity | Outcome (share / margin) | Confirming KPI |
|---|---|---|---|---|---|
| Core EDA, US-led AI/HPC | Direct, both | Parity (inferred; no price data) | High, both | Cadence share up, margin up; Synopsys share flat to down, margin up through cost cuts | Organic core EDA growth gap, FY2027 |
| Silicon IP | Direct, both | Synopsys at or above on breadth; Cadence winning on PPA (CEO claim) | Synopsys disrupted 2025; Cadence high | Cadence share up; Synopsys share stabilizing, margin recovering from 24% | Synopsys IP adj. margin back above 30% |
| Simulation and system analysis | Synopsys direct plus channel; Cadence mostly direct | Synopsys above as category leader | Synopsys high; Cadence integrating D&E | Synopsys share flat to up, margin up on synergies; Cadence share up from a small base, margin down in 2026 | Ansys growth at 10%+ in FY2027 |
The tension sits in the IP row: Cadence is growing fastest exactly where Synopsys has the larger installed catalogue, which says the 2025 losses were about execution, and execution can be fixed.
1. Mix (Major). IP was 25% of Synopsys revenue against 14% at Cadence in FY2025; IP is people-intensive and recognized per design, which lowers gross margin and adds volatility. Closing it means growing EDA and simulation faster than IP for years; the Ansys deal is that move, already made. 2. Organizational load (Moderate, inferred). Since mid-2025 Synopsys has closed Ansys, sold three businesses, cut 10% of staff, settled with an activist (Elliott, May 2026) and is replacing its general counsel, while Cadence's 2026 integration is a $2.9B add-on. This closes by itself once integration ends, likely through FY2027. 3. China mix (Moderate). Synopsys's China book is IP-heavy and that is where local substitution happens first (Q1 FY2026 call); this does not close with money.
Both report under US GAAP by function, so no by-nature mapping is needed. Three adjustments keep the comparison honest. Cost of revenue is shown before acquired-intangible amortization, because Ansys added $204M of it to Synopsys's FY2025 cost of revenue and would otherwise swamp the ratio (both companies disclose the amount). Restructuring, opex amortization and Cadence's $128.5M 2025 DOJ/BIS loss sit on separate lines and are excluded from R&D and SG&A. Synopsys's FY2025 G&A includes part of $254.8M of acquisition and divestiture costs, about 3.6 points of revenue, which the filings do not split by line. Years: Synopsys FY2023 to FY2025 (to October, continuing operations, FY2024 had 53 weeks), Cadence 2023 to 2025 (to December).
| % of sales, 3-year average | Synopsys | Cadence | Gap | What drives it |
|---|---|---|---|---|
| COGS* | 18.6% | 11.5% | +7.1 | IP implementation services and maintenance cost at Synopsys |
| R&D | 34.6% | 34.0% | +0.6 | The entry ticket to a three-vendor oligopoly; near-identical |
| SG&A | 23.4% | 22.0% | +1.4 | Ansys channel selling and deal costs in FY2025; Synopsys was lower in FY2023 |
| Gross margin* | 81.4% | 88.5% | -7.1 | Mix: IP is a quarter of Synopsys revenue vs 14% at Cadence |
| GAAP EBIT margin | 19.7% | 29.3% | -9.6 | Gross margin gap plus Ansys amortization and deal costs in FY2025 |
*Before acquired-intangible amortization in cost of revenue. Sources: Synopsys FY2024 and FY2025 Q4 results releases and FY2025 10-K; Cadence FY2024 and FY2025 Q4 results releases and 10-Ks.
| Synopsys, % of sales | COGS* | R&D | SG&A | Gross margin* | GAAP EBIT | Non-GAAP op. margin |
|---|---|---|---|---|---|---|
| FY2023 | 18.5 | 34.8 | 20.7 | 81.5 | 23.9 | 36.3 |
| FY2024 | 18.6 | 34.0 | 23.3 | 81.4 | 22.1 | 38.5 |
| FY2025 | 18.6 | 35.1 | 26.1 | 81.4 | 13.0 | 37.3 |
| Cadence, % of sales | COGS* | R&D | SG&A | Gross margin* | GAAP EBIT | Non-GAAP op. margin |
|---|---|---|---|---|---|---|
| 2023 | 9.6 | 35.3 | 22.8 | 90.4 | 30.6 | 42.0 |
| 2024 | 12.7 | 33.4 | 22.2 | 87.3 | 29.1 | 42.5 |
| 2025 | 12.4 | 33.4 | 21.1 | 87.6 | 28.2 | 44.6 |
Synopsys non-GAAP for FY2023 is total segment adjusted operating margin ($1,928.0M on $5,318.0M); FY2024 and FY2025 as reported in the supplement. Stock compensation: Synopsys 9.6%, 10.7%, 12.7% of revenue; Cadence 8.0%, 8.4%, 8.6% (10-Ks).
Gross margin is the persistent gap: about 7 points in each of the three years, before any Ansys amortization. The mechanism comes from Step 1: a quarter of Synopsys's revenue was Design IP, which its CFO calls "such a people-intensive body of work to deliver" that IP margins always sit below the corporate average (Q4 FY2025 call), and Synopsys books more maintenance and service revenue at lower margin. R&D intensity is the same at both companies, so the gap is not an innovation shortfall; it is mix and scale per employee, with Cadence producing about $420K of revenue per employee on its 2026 guide against about $384K at Synopsys (inferred from guides and headcount). The gap confirms the power map only partly: it reflects what Synopsys sells, not weaker pricing. That is why the margin call is narrow, since Synopsys's plan to grow EDA and simulation faster than IP, and to cut costs, attacks the mechanism directly.
| Metric | Threshold | By when | If it hits, it favors | Where published |
|---|---|---|---|---|
| Cadence backlog | At least $8.5B, with 55%+ due in 12 months | Q4 2026 results, Feb 2027 | Cadence | Cadence release and 10-K |
| Synopsys Design IP growth | At least 15% year on year, adj. margin 30%+ | Q2 FY2027, May 2027 | Synopsys | Synopsys 10-Q segment note |
| Synopsys non-GAAP operating margin | At least 44% for FY2027 | Dec 2027 | Synopsys | Synopsys Q4 FY2027 release |
| Organic core EDA growth gap (Cadence minus Synopsys) | 3 points or less favors Synopsys; 5+ favors Cadence | FY2027, by Feb 2028 | Either | 10-Ks and calls |
| Cadence China revenue | No worse than -15% year on year in 2027 | Quarterly through 2027 | Cadence | Cadence 10-Q geography note |